The Oxford Handbook of Danish Politics 9780198833598, 0198833598

Table of contents :
Cover
The oxford handbook of DANISH POLITICS
Copyright
Contents
List of Contributors
Preface
List of Figures
List of Tables
Chapter 1: Why a Book on Danish Politics?
Structure of the Handbook
Part I: POLITY
Chapter 2: The Constitution
The Constitutional Act as the Framework for Danish Democracy
The History of the Constitutional Act
The King and the Monarchy
The Folketinget
The Legislative Process
Constitutional Provisions on Parliamentarism
Forming a Government
Parliamentary Oversight of Government and Ministers
Denmark’s International Relations and Delegation of Sovereignty
The Courts
Constitutional Rights
The Constitution that Never Changes
References
Chapter 3: Greenland, the Faroe Islands, and Denmark: Unity or Community?
From Margins to Centre Stage
A Constitutional Oxymoron
Competencies Reaching into Core State Functions
Seeking Equality in Asymmetrical Relations
No Turning Back after the Second World War
Finding Inspiration for New Equalities
Eligible for Equal Sovereignty?
Geopolitical and Paradiplomatic Sovereignty Games
Imperial and Cold War Sovereignty Games
European Sovereignty Games
Arctic Sovereignty Games
Conclusion
References
Chapter 4: The Monarch: Head of State and National Symbol
History
Succession
The Royal House
The Court
The Chapel of the Royal Orders of Knighthood
The Monarch’s Functions
The Legitimacy of the Monarchy
Conclusion
References
Chapter 5: The Electoral System: Fair and Well-Functioning
The Structure of the Electoral System
How the Electoral System Actually Works
Step 1
Step 2
Step 3
Steps 4 and 5
Step 6
The Nature of Electoral Law Amendments
Comparisons with Other Electoral Systems
Conclusion
References
Chapter 6: Electoral Turnout: Strong Social Norms of Voting
Very High and Stable Turnout
Who Votes?
Denmark: A European Frontrunner in Get-Out-The-Vote (GOTV)
Conclusion
References
Chapter 7: The Parliament (FOLKETINGET): Powerful, Professional, and Trusted?
A Powerful Parliament?
Legislative Influence
Ex Ante Selection Control
Ex Post Oversight Control
A Parliament with Powerful Institutions
A Professional Parliament?
Institutional Resources
Specialization and Expertise
Politics as a Profession
A Professional Parliament with Lower Specialization
A Trusted Parliament?
Conclusion
References
Chapter 8: The Government and the Prime Minister: More than Primus Inter Pares?
A Popular Case of Coalitions and Minority Governments
Government Formation and Termination in Denmark
Coalitions and Coalition Governance
Government Termination
The Prime Minister
Ministers and Ministerial Turnover
Coordination within the Government
Conclusion
References
Chapter 9: Organizing Central Government: A Pragmatic Meritocracy?
What Does Central Government Do?
Who Runs Central Government?
The Organization of Central Government
How Is Central Government Coordinated?
The Bureaucracy
The Politicization of the Bureaucracy
The Ethics of Bureaucracy
Keeping Central Government in Check
Conclusion
Note
References
Chapter 10: Municipalities and Regions: Approaching the Limit of Decentralization?
Local Politics in the Eye of Decentralization, a Big-Bang Local Government Reform, and a Toughened National Spending Regime
The Administrative Structure of Denmark
The Local Political System and Democracy
The Administrative Structure of Municipalities and Regions
Local Discretion in the Light of Changed Central-Local Relations
Fiscal Autonomy Regarding Expenditures and Revenues
Expenditures
Revenues
Challenges and Opportunities
Conclusion
References
Chapter 11: Corporatism: Exaggerated Death Rumours?
Corporatism Defined: A Variety of Democracy
The Origin of Danish Corporatism
Declining Corporatism?
Committee Representation
Studies of Government Reform Strategies
Why Is Corporatism so Tenacious?
Conclusion
References
Chapter 12: Corruption and Bureaucratic Reforms‘: Getting to Denmark’?
‘Getting to Denmark’
Corruption and Anti-Corruption in Denmark
Historical Bureaucratic Reforms and the Fight Against Corruption in Denmark
Steps in the Establishment of Rule of Law
An Institutions Matter Model
Conclusion
References
Chapter 13: Coordination for the European Union: A Strong and Stable Institution
The Coordination System for the European Union
The Diachronic Perspective: The Evolution of the Danish Coordination System from 1950–2019
The Synchronic Perspective: The Organization of the Danish EU Coordination System
The Evaluative Perspective: The Weaknesses and Strengths of the Danish EU Coordination System
The Comparative Perspective: Contrasting EU Coordination in Denmark with Other Member States
Conclusion
Acknowledgement
Note
References
Part II: POLITICS
Chapter 14: The Party System: Open yet Stable
The Earthquake Election and its Long-Term Aftermath
The Danish Party System
The Four Old Parties and the Historical Cleavages in the Danish Party System
The Development of the Danish Party System since 1973
After the Turn of the Century
The Political Dimensions of the Current Danish Party System
Electoral Support
Party Organizations
Party Membership
Intra-Party Democracy
Party Financing
Conclusion
Note
References
Chapter 15: Governments in Action: Consensual Politics and Minority Governments
Party Unity
The Stability of Danish Minority Governments
How Do Governments in Denmark Build Legislative Coalitions?
Patterns of Legislative Behaviour
Legislative Agreements
Conclusion
Notes
References
Chapter 16: Classes and Politics: A Changing Relationship
Class and Danish Politics in a Historical Perspective
Theories of Class and Politics
Class Relations and Class Identity in Denmark
The Development of Class-Based Party Choice
Conclusion
References
Chapter 17: Danish Public Opinion: Stability, Change, and Polarization
Stability or Change in Danish Public Opinion
Economic Redistribution
Immigration
The Environment
The EU
Social Polarization of Public Opinion over Time
Attitudes Toward Redistribution
Immigration
The Environment
The EU
Partisan Polarization of Public Opinion over Time
Attitudes toward Redistribution
Immigration
The Environment
The EU
Danish Public Opinion in a Comparative Perspective
Economic Redistribution
Immigration
The Environment
The EU
Conclusion
References
Chapter 18: The Social Democratic Party: From Exponent of Societal Change to Pragmatic Conservatism
Historical Overview
The Foundation and the Organizational Period, 1871–1901
The Integration into Danish Parliamentarianism, 1901–1929
Government Responsibility and Defence of Democracy, 1929–1947
The Welfare State and the Cold War, 1947–1989
Globalization and the Neoliberal Turn, 1989–2020?
Present-Day Policy Positions and Issue Emphasis
The Welfare State
The Private Sector
Education
European Policy and Foreign Affairs
Immigration Policies
Globalization
Voters
Party Organization and Party Democracy
Conclusion
References
Chapter 19: The Liberal Party: From Agrarian and Liberal to Centre-Right Catch-All
The Liberal Party
Brief Historical Outline
The Electorate of the Liberals: Development and Composition
The Ideology and Positions of the Liberals
Party Members, Organization, and Finance
Relations to Other Political Parties
What Policy Difference Have the Liberals Made after 2001?
Conclusion
Notes
References
Chapter 20: The Danish People’s Party: Centre-Oriented Populists?
A Star is Born: The Creation of the Danish People’s Party
Electoral Support
Government Participation
Party Policy and Ideology
Political Dimensions
Policies and Issues
Populist Radical Right?
The Impact of the Danish People’s Party
One for All, All for One: Party Discipline and Party Leadership
Party Discipline
Party Leadership
Party Organizational Resources
Conclusion
References
Chapter 21: The Red-Green Alliance: Is it Red or Green?
Introducing the Red-Green Alliance
Ideology
Voters of the Red-Green Alliance
Issue Positions of the Red-Green Alliance
Issue Priorities of the Red-Green Alliance (I): The Environment and Labour Market Policy
Issue Priorities of the Red-Green Alliance (II): The EU and Foreign Aid
Party Organization of the Red-Green Alliance
Party Finances of the Red-Green Alliance
Red-Green Alliance Party Membership
Red-Green Alliance Intra-Party Democracy
Conclusion
Notes
References
Chapter 22: Dynamic Stability: The Anchors of Voting Behaviour
Increasing Voter Volatility—But Not at the Bloc Level
Theories about the Anchors of Voting Behaviour
Traditional Cleavage-Based Anchors: Place of Residence and Class
New Socio-Structural Anchors: Education and Gender
A Two-Dimensional Ideological Anchor: Economic Redistribution and Cultural Values
A Changing Political Space: Party-Voter Dynamics
Conclusion
References
Chapter 23: Gender and Politics: The Limits of Equality Politics
Setting the Stage
Framing the Chapter
The Evolution of Theoretical Debates and Key Concepts
Scandinavian Gender Equality and Citizenship
Gender and Political Mobilization, Participation, and Representation
Gender Equality Policies
Conclusion
Note
References
Chapter 24: Local Elections: Localized Voting Within a Nationalized Party System
Second-Order or Second-Tier Elections?
How Many Turn Out at the Local Polls?
How ‘Local’ Is the Local Vote?
Is the Local Party System a Copy of the National Party System?
Still More Men than Women?
Who Performs Local Political Leadership?
Conclusion
References
Chapter 25: Referendums in Denmark: Influence on Politics
Referendums in Denmark
The Rules for Referendums in Denmark
Why Are (EU) Referendums Convened?
Do Campaigns Matter?
Voter Behaviour: Issue Voting Versus Second-Order Elections
The 1992/1993 Referendums
No Votes in Opt-Out Referendums: The 2015 No Vote
The Effects of EU Referendums
Internal Effects: Keeping the EU Out of Domestic Politics
External Effects: Opt-Outs (and Leaving the Union)
Internal and External Effects: No Votes and Greenlandic Exit from the EU
Conclusion
Notes
References
Chapter 26: Media and Politics: The Danish Media System in Transformation?
Democracy and the Crisis of Journalism
The Danish Media System in a Comparative Perspective
The Media-Politics Nexus
The Crisis of Journalism
Conclusion
References
Chapter 27: Interest Groups: A Democratic Necessity and a Necessary Evil
Interest Groups: Definitions and Distinctions
The Origin and Development of the Interest Group System
The Rise of New Social Movements and Citizen Groups
Interest Groups as Associations: Members, Resources, and Organization
The Political Role of Interest Groups
New Actors: Think Tanks and Public Affairs Bureaus
Conclusion
References
Part III: POLICIES
Chapter 28: In War and Peace: Security and Defence Policy in a Small State
The Willing but Pragmatic Activist
The Shadow of 1864: Two Danish Traditions on War and Peace
The Shame of 1945: From De-Militarized Neutral to Allied with Reservations
The Solution of 1989: Denmark’s Military Activism
Conclusion
Notes
References
Chapter 29: Foreign Policy: New Directions in a Changing World Order?
A Variety of Approaches to Studying Danish Foreign Policy
Discourse Analysis and Danish Foreign Policy
Discourses and Foreign Policy after the Cold War
The EU as the Key Platform for Danish Foreign Policy
The Transatlantic Link
Activism
Danish Foreign Policy in a Changing World
Conclusion
Notes
References
Chapter 30: Danish European Union Policies: Sailing Between Economic Benefits and Political Sovereignty
Varieties of EU Policies
Trade Policy
Common Agricultural Policy
The Internal Market
Opt-Outs: The Euro, Justice and Home Affairs, and Defence Policy
Conclusion
Acknowledgements
Notes
References
Chapter 31: Development Policy From Consensus to Contention?
Policy Changes in the New Millennium
Changes in Danish Development Policy
Size and Composition of the Aid Budget
The Danish Development Strategy Papers and Their Thematic Focus
Factors Behind the Changing Development Policy
Denmark Reacted to International Trends
Domestic Factors Before 2001
Domestic Factors after 2001
Conclusion
Note
References
Chapter 32: Economic Policy: From Disequilibrium to Flexicurity
A Small Open Economy with an Extended Welfare State
Economic Developments
The 1970s: Disequilibria Build Up
The 1980s: The Big Upturn and the Seven Meagre Years
The Exchange Rate Peg
Current Account Problems and Savings Shortage
Tax Reforms
The 1990s: Founding the Flexicurity Model
Flexicurity Reforms
2000–2018: Overheating and the Financial Crisis
Fiscal Policy
Testing the Flexicurity Model
Ageing, Fiscal Sustainability, and Reforms Directed at Labour Supply
Conclusion
Notes
References
Chapter 33: Welfare State Policies: From the Beginning Towards an End?
Characteristics of the Danish Welfare Model
Historical Developments up to the 1980s
New Challenges to the Danish Welfare State
Restructuring and Reforming the Danish Welfare State from the 1980s
Welfare Cutbacks and Expansions
A Strong Focus on Getting More People into the Workforce
The Challenge of an Ageing Population
Immigration and Integration of Immigrants as a New and Highly Politicized Policy Area
Public Sector Reforms and Strict Control of Expenditure Levels
Conclusion
Notes
References
Chapter 34: The Danish Labour Market Model: Is the Bumblebee Still Flying?
The Danish Labour Market Model and Labour Market Policies
The Creation of the Stable Institutions of the Danish Labour Market Model of Regulation
The Establishment of the Danish Ghent System
The Development of Active Labour Market Policy
The Components of the Danish Labour Market Model
New Employment Policies and Reforms of the Ghent System in the New Millennium
Are the Stable Institutions of the Danish Labour Market Model Eroding?
Conclusion
References
Chapter 35: Education Policy: Power, Conflict, and Cooperation
The Regrettable Status of Research on Danish Education Politics
The Restructuring of the Danish School System along Egalitarian Lines, 1903–1958
Comprehensive Education Consolidated, 1958–1975
Market-Led Reforms of Danish Education, 1982–2014
School Choice and Private Schools
PISA Shock
The Teachers’ Union Reacts: Lock-Out
Enhancing Academic Standards Further: The 2014 School Act
Conclusion
References
Chapter 36: Health Policy: The Submerged Politics of Free and Equal Access
Consensus about Universal Healthcare
Three Dimensions of the Danish Healthcare System
Financing: Expansion of Access and Tax-Financed Health Insurance
Regulation and Medical Governance
Delivery: Public and Private Provision of Services in the Danish Healthcare System
Health Politics: Party Politics and Competition for Issue Ownership
Conclusion
Note
References
Chapter 37: Immigration and Immigrant Integration Policy: Public Opinion or Party Politics?
The Shattered Image of Denmark
Danish Immigration History and Policy Development
Anti-Immigrant Danes?
Political Dynamics and Politicization
Policy and Politicization Effects
Conclusion
References
Chapter 38: Agricultural and Fisheries Policy: Towards Market Liberalism
Market Orientation of Agricultural and Fisheries Policies
Market Liberalism
Water, Carbon Emissions, and Environmental Policies: Cost-Effective Regulation
Structural Development Policy: From Restrictions to Market Liberalism
Organic Food Policy: Actively Developing Markets
Fisheries Policy: From Almost Open Access to Property Rights and Market-Based Management
Conclusion
Notes
References
Chapter 39: Environment, Energy, and Climate Policy: From Energy Supply to Climate Gases
Denmark: A Green Frontrunner Nation?
Environmental Policy
Environmental Problems in Denmark: The Special Role of Agriculture
The Organization of Environmental Administration
Key Principles and Instruments in Danish Environmental Policy
The Build-Up Phase of Environmental Regulation, 1971–1987
The Heyday of Environmental Policy, 1987–2002
The Increasing Role of the EU since 2002
The Evolution of Environmental Quality: Has Environmental Policy Worked?
Recent Bones of Contention in Environmental Policy
Energy and Climate Policy
1974–1989: Energy Security as the Overriding Concern
1990–2001: Climate Policy Takes Precedence
Since 2002: Stop-Go Policy
The Outcome: Energy Production and Consumption Since the Early 1970s
The Evolution of Renewable Energy
Conclusion
Acknowledgements
References
Chapter 40: Research Policy: Transformations and Tensions
The Social Contract of Science
Competing Conceptions of Research Quality and Societal Impact
Broad Trends in Danish Research Policy
Transformation of the Research Funding System and New Conceptions of the Universities’ Third Mission
Transformation of the Public Research Funding System
New University Third Mission Demands
The Two Cases in a Comparative Perspective
Conclusion
References
Part IV: POSTSCRIPT
Chapter 41: Polity, Politics, and Policies: A Reconsideration
Polity
Basic State Institutions: Aged but Still Functioning Quite Well
Basic Political Institutions: Generally Well-Functioning and Supported by Trust and Civic Norms
Administrative Institutions
Politics
Party System, Minority Governments, Classes, and Long-Term Voting
Four Political Parties and the Anchors of Voting Behaviour: The Social Democratic Party, the Liberal Party, the Danish People’s Party, and the Red-Green Alliance
Gender, Local Elections, Referendums, Media, and Interest Groups
Policy
Looking to the Outside World: Security and Defence, New World Order, the EU, and Development Policy
The Welfare State Machinery: Economic Policy, the Welfare State, the Labour Market, Education, Health, and Immigration
Natural Resources and Know-How: Agriculture and Fisheries, Environment and Climate Policy, and Research Policy
Still Worth Getting to Denmark?
References
Index

Citation preview

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

T h e Ox f o r d H a n d b o o k o f

DA N ISH POL I T IC S

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

the oxford handbook of

DANISH POLITICS Edited by

PETER MUNK CHRISTIANSEN, JØRGEN ELKLIT, and

PETER NEDERGAARD

1

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

1 Great Clarendon Street, Oxford, ox2 6dp, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2020 The moral rights of the authors have been asserted First Edition published in 2020 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2020937356 ISBN 978–0–19–883359–8 Printed and bound by CPI Group (UK) Ltd, Croydon, cr0 4yy Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

Contents

List of Contributorsix Prefacexiii List of Figuresxv List of Tablesxix

1. Why a Book on Danish Politics?

1

Peter Munk Christiansen, Jørgen Elklit, and Peter Nedergaard

PA RT I .   P OL I T Y Se cti on E di tor: Jørgen E lk lit

2. The Constitution

9

Jens Peter Christensen

3. Greenland, the Faroe Islands, and Denmark: Unity or Community?28 Ulrik Pram Gad

4. The Monarch: Head of State and National Symbol

46

Jes Fabricius Møller

5. The Electoral System: Fair and Well-Functioning

56

Jørgen Elklit

6. Electoral Turnout: Strong Social Norms of Voting

76

Kasper M. Hansen

7. The Parliament (Folketinget): Powerful, Professional, and Trusted?

88

Helene Helboe Pedersen

8. The Government and the Prime Minister: More than Primus Inter Pares?107 Martin Ejnar Hansen

9. Organizing Central Government: A Pragmatic Meritocracy? Caroline Howard Grøn and Heidi Houlberg Salomonsen

124

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

vi   contents

10. Municipalities and Regions: Approaching the Limit of Decentralization?

141

Kurt Houlberg and Niels Ejersbo

11. Corporatism: Exaggerated Death Rumours?

160

Peter Munk Christiansen

12. Corruption and Bureaucratic Reforms: ‘Getting to Denmark’?

177

Mette Frisk Jensen and Gert Tinggaard Svendsen

13. Coordination for the European Union: A Strong and Stable Institution

193

Mads Dagnis Jensen and Peter Nedergaard

PA RT I I .   P OL I T IC S Se cti on E di tor : P eter Mu nk C hristiansen

14. The Party System: Open yet Stable

213

Christoffer Green-Pedersen and Karina Kosiara-Pedersen

15. Governments in Action: Consensual Politics and Minority Governments230 Christoffer Green-Pedersen and Asbjørn Skjæveland

16. Classes and Politics: A Changing Relationship

242

Gitte Sommer Harrits and Rune Stubager

17. Danish Public Opinion: Stability, Change and Polarization

259

Peter Thisted Dinesen, Rune Slothuus, and Rune Stubager

18. The Social Democratic Party: From Exponent of Societal Change to Pragmatic Conservatism

278

Rasmus Mariager and Niels Wium Olesen

19. The Liberal Party: From Agrarian and Liberal to Centre-Right Catch-All296 Flemming Juul Christiansen

20. The Danish People’s Party: Centre-Oriented Populists?

313

Karina Kosiara-Pedersen

21. The Red-Green Alliance: Is it Red or Green? Henrik Bech Seeberg and Ann-Kristin Kölln

329

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

contents   vii

22. Dynamic Stability: The Anchors of Voting Behaviour

347

Kasper M. Hansen and Rune Stubager

23. Gender and Politics: The Limits of Equality Politics

365

Christina Fiig and Birte Siim

24. Local Elections: Localized Voting Within a Nationalized Party System

382

Ulrik Kjær

25. Referendums in Denmark: Influence on Politics

400

Derek Beach

26. Media and Politics: The Danish Media System in Transformation?417 Thomas Olesen

27. Interest Groups: A Democratic Necessity and a Necessary Evil

433

Anne Skorkjær Binderkrantz

PA RT I I I .   P OL IC I E S Se cti on E di tor : P eter Nedergaard

28. In War and Peace: Security and Defence Policy in a Small State

453

Anders Wivel

29. Foreign Policy: New Directions in a Changing World Order?

470

Henrik Larsen

30. Danish European Union Policies: Sailing Between Economic Benefits and Political Sovereignty

487

Mads Dagnis Jensen and Peter Nedergaard

31. Development Policy: From Consensus to Contention?

502

Anne Mette Kjær

32. Economic Policy: From Disequilibrium to Flexicurity

521

Torben M. Andersen

33. Welfare State Policies: From the Beginning Towards an End? Klaus Petersen

540

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

viii   contents

34. The Danish Labour Market Model: Is the Bumblebee Still Flying?

559

Laust Høgedahl

35. Education Policy: Power, Conflict, and Cooperation

577

Susanne Wiborg

36. Health Policy: The Submerged Politics of Free and Equal Access

592

Lars Thorup Larsen

37. Immigration and Immigrant Integration Policy: Public Opinion or Party Politics?

609

Kristina Bakkær Simonsen

38. Agricultural and Fisheries Policy: Towards Market Liberalism

627

Carsten Daugbjerg, Peder Andersen, Henning Otte Hansen, and Brian H. Jacobsen

39. Environment, Energy, and Climate Policy: From Energy Supply to Climate Gases

644

Peter Birch Sørensen

40. Research Policy: Transformations and Tensions

664

Kaare Aagaard and Niels Mejlgaard

PA RT I V.   P O S T S C R I P T 41. Polity, Politics, and Policies: A Reconsideration

685

Peter Munk Christiansen, Jørgen Elklit, and Peter Nedergaard

Index

701

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

List of Contributors

Kaare Aagaard  Senior Researcher, Department of Political Science, Aarhus University Peder Andersen  Professor Emeritus, Department of Food and Resource Economics, University of Copenhagen Torben M. Andersen  Professor, Department of Economics and Business Economics, Aarhus University Derek Beach  Professor, Department of Political Science, Aarhus University Anne Skorkjær Binderkrantz  Professor, Department of Political Science, Aarhus University Jens Peter Christensen  Supreme Court Justice Flemming Juul Christiansen  Associate Professor, Department of Social Sciences and Business, Roskilde University Peter Munk Christiansen  Professor, Department of Political Science, Aarhus University Carsten Daugbjerg  Professor, Department of Food and Resource Economics, University of Copenhagen Peter Thisted Dinesen  Professor, Department of Political Science, University of Copenhagen Niels Ejersbo  Deputy Director, University College Copenhagen Jørgen Elklit  Professor Emeritus, Department of Political Science, Aarhus University Christina Fiig  Associate Professor, School of Culture and Society, Aarhus University Ulrik Pram Gad  Associate Professor, Department of Culture and Learning, Aalborg University Christoffer Green-Pedersen  Professor, Department of Political Science, Aarhus University Caroline Howard Grøn  Associate Professor, Department of Political Science, Aarhus University Henning Otte Hansen  Senior Adviser, Department of Food and Resource Economics, University of Copenhagen

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

x   list of contributors Kasper M. Hansen  Professor, Department of Political Science, University of Copenhagen Martin Ejnar Hansen  Senior Lecturer, Department of Social and Political Sciences, Brunel University London Gitte Sommer Harrits  Vice Rector Academic, VIA University College Laust Høgedahl  Associate Professor, Department of Politics and Society, Aalborg University Kurt Houlberg  Professor, The Danish Center for Social Science Research Brian H. Jacobsen  Associate Professor, Department of Food and Resource Economics, University of Copenhagen Mads Dagnis Jensen  Associate Professor, Department of International Economics, Government and Business, Copenhagen Business School Mette Frisk Jensen  Senior Advisor, School of Culture and Society, Aarhus University Anne Mette Kjær  Professor, Department of Political Science, Aarhus University Ulrik Kjær  Professor, Department of Political Science and Public Management, University of Southern Denmark Ann-Kristin Kölln  Associate Professor, Department of Political Science, Aarhus University Karina Kosiara-Pedersen  Associate Professor, Department of Political Science, University of Copenhagen Henrik Larsen  Professor, Department of Political Science, University of Copenhagen Lars Thorup Larsen  Associate Professor, Department of Political Science, Aarhus University Rasmus Mariager  Associate Professor, The SAXO-Institute, University of Copenhagen Niels Mejlgaard  Professor, Department of Political Science, Aarhus University Jes Fabricius Møller  Associate Professor, The SAXO-Institute, University of Copenhagen Peter Nedergaard  Professor, Department of Political Science, University of Copenhagen Niels Wium Olesen  Associate Professor, Department of History and Classical Studies, Aarhus University Thomas Olesen  Professor, Department of Political Science, Aarhus University Helene Helboe Pedersen  Professor, Department of Political Science, Aarhus University

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

list of contributors   xi Klaus Petersen  Professor, Danish Centre for Welfare Studies & Danish Institute of Advanced Studies, University of Southern Denmark Heidi Houlberg Salomonsen  Associate Professor, Department of Management, Aarhus University Henrik Bech Seeberg  Associate Professor, Department of Political Science, Aarhus University Birte Siim  Professor Emeritus, Department of Politics and Society, Aalborg University Kristina Bakkær Simonsen  Associate Professor, Department of Political Science, Aarhus University Asbjørn Skjæveland  Associate Professor, Department of Political Science, Aarhus University Rune Slothuus  Professor, Department of Political Science, Aarhus University Peter Birch Sørensen  Professor, Department of Economics, University of Copenhagen Rune Stubager  Professor, Department of Political Science, Aarhus University Gert Tinggaard Svendsen  Professor, Department of Political Science, Aarhus University Susanne Wiborg  Reader, Institute of Education, University College London Anders Wivel  Professor, Department of Political Science, University of Copenhagen

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

Preface

For a considerable while, there has been a need for an up-to-date English-language book on Danish politics and Danish political institutions. Such a book should aim at meeting the needs of both established researchers and students abroad—and in Denmark—and hopefully also the general public. A project to this effect materialized when Peter Nedergaard in 2017 suggested to Peter Munk Christiansen that the time was ripe for a Handbook of Danish Politics. Jørgen Elklit joined the editorial team subsequent to that. Oxford University Press immediately responded very positively to our book proposal, and the reviewers’ comments were very encouraging. We would like to thank Dominic Byatt from Oxford University Press for his speedy and constructive handling of our project from beginning to end. The book idea also met positive responses when we approached potential authors. We even had requests from well-qualified academics to be part of the book project when the rumours about our plans spread in the relevant circles. We accepted a few of these requests. We thank all authors for their contributions and for productive deliberations during the editorial process. Many of the early chapter drafts were reviewed and debated at the Annual Meeting of the Danish Political Science Association at Hotel Vejlefjord in October 2018. We thank the organizers for making room for us at the meeting. Several of the chapters benefitted substantially from this early presentation and discussion. Many of the authors have also received various kinds of help and advice from colleagues, civil servants in the ministries, the Danish Parliament, from organisations, and so on. This is much appreciated, and the book has certainly benefitted from such help and advice. In the final scrutiny of chapter manuscripts, Natasha Elizabeth Perera and Mette Aagaard, Department of Political Science, Aarhus University, were instrumental in ensuring that manuscripts were completed in a satisfactory way. This involved improvement of authors’ written English, general sub-editing, control of references, and indexing. We are grateful as well as very impressed by Natasha’s and Mette’s professionalism and efficiency. We would also like to thank Rob Wilkinson for copy-editing the entire book. We are also grateful for the funding and support for language editing, and so on, from the departments of political science at the University of Copenhagen and Aarhus University.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

xiv   preface We hope that The Oxford Handbook on Danish Politics has become what we from the outset intended it to be—namely a scholarly book about most aspects of Danish politics that is also accessible to the general public. Peter Munk Christiansen, Jørgen Elklit, and Peter Nedergaard

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

List of Figures

3.1 When seeing Greenland and the Faroe Islands between North America, Scandinavia, and Eurasia in polar projection, Denmark disappears 

29

5.1 Effective number of parties in the electorate, N(v), and in Parliament, N(s), 1918–2019

62

5.2 The six steps of seat allocation in the Danish electoral system

64

5.3 2019 ballot paper from Nomination District 10 in Copenhagen MMC

70

6.1 Turnout in Denmark 1979–2019, per cent

77

6.2 The rollercoaster: Turnout across age and gender in the 2015 parliamentary election and the 2017 local elections

79

6.3 Turnout across ethnicity and age in the 2015 parliamentary elections and the 2017 local elections

80

7.1 Number of government and Private Member Bills and laws, 1973–2018

91

7.2 Number of Section 20 Questions and Committee Questions, 1973–2018

94

7.3 Parliamentary group subsidy and staff, 2001–2017

97

7.4 Age and education of Danish MPs, 1953–2019

99

7.5 Trust in parliament, government, and political parties, 2008–2019 (per cent)

101

10.1 Municipal service expenditures, 2007–2019 (2019 price and task level)

149

11.1 Yearly number of commission reports, 1972–2018

166

11.2 Share of commissions with at least one civil servant, expert, or interest group as member, 1972–2018

166

11.3 Experts’, civil servants’, and interest groups’ share of commission seats, 1972–2018166 12.1 An ‘institutions matter’ model

186

13.1 The system of Danish EU coordination

200

14.1 Party positions on the economic left-right and new politics dimensions 2014

219

14.2 Electoral support of parliamentary parties as of 2019

220

14.3 Danish party membership figures and member/electorate ratios

222

14.4 Danish party membership figures, 2000–2018

223

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

xvi   list of figures 14.5 Danish party funding 2015

224

16.1 Development of class identification (including ‘leaners’)

247

16.2 Occupational class and support for centre-left parties

250

16.3 Sector and support for centre-left parties

250

16.4 Income and support for centre-left parties

251

16.5 Education and support for centre-left parties (Higher Education vs others) 252 16.6 Predictive power of class variables on support for centre-left parties

253

16.7 Class and education, Alford Index Values

254

17.1 Development in public opinion toward economic redistribution, immigration, the environment, and the EU over time

261

17.2 Socio-demographic differences in attitudes toward economic redistribution, per cent agreeing that income differences are too large

264

17.3 Socio-demographic differences in attitudes toward immigration, per cent agreeing that immigration is a threat

265

17.4 Socio-demographic differences in attitudes toward the environment, per cent agreeing that economic growth goes before the environment

266

17.5 Socio-demographic differences in attitudes toward the EU, per cent positive267 17.6 Development in opinions on economic redistribution, immigration, the environment, and the EU over time by partisan groups

269

17.7 Public opinion on four issues in Denmark, Sweden, Norway, Germany, Great Britain, Poland, and Spain, 2003–2015, per cent in agreement

273

19.1 Share of farmers in Denmark of total labour force, and average voter support for the Liberals for each decade since the 1920s, percentages

298

19.2 Liberal Party members, 1930–2015, five-year intervals

303

20.1 Electoral support of the Progress Party and the Danish People’s Party, 1973–2019315 20.2 Left-Right placement of the Danish People’s Party, 1999–2014

318

21.1 Vote share and seats for the Red-Green Alliance

331

21.2 Number of questions to the Minister per year by parties in the Danish Parliament, 1994–2012

332

21.3 Vote share to the Red-Green Alliance across constituencies in the Danish 1994 and 2015 national elections

335

21.4 Average attention to issue areas in press releases for the Red-Green Alliance, Socialist People’s Party, and the Social Democrats in Denmark, 2004–2017

338

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

list of figures   xvii 21.5 The Red-Green Alliance’s attention to six issues in their press releases, 2004–2017339 21.6 Annual party income (Central Office), 1990–2016

340

21.7 Distribution of sources of party income in 2016

341

21.8 Membership size as share of the entire electorate, 1990–2016

342

22.1 Electoral volatility, 1971–2019

348

22.2 Support for the left bloc by place of residence and class, 1971–2019

352

22.3 Support for the left bloc by education and gender, 1971–2019

354

22.4 Correlation between economic redistribution and left-right selfplacement and between cultural values and left-right self-placement, 1979–2019356 22.5 Support for the left bloc by taxation and immigration, 1979–2019

358

22.6 Parties’ positions on taxation and immigration, 1981, 1990, and 2019

360

24.1 The nationalization of the party system, 1966–2017 (average for all municipalities)390 24.2 Female representation at the local, regional, and national level, 1966–2017

392

27.1 Membership activity and influence

440

27.2 The use of different frames in Denmark and the UK, percentages

443

28.1 Danish defence expenditure 1950–1989 as percentage of Gross Domestic Product (GDP)

458

28.2 Denmark’s military contributions to international operations

460

28.3 Danish defence expenditure 1990–2023 as percentage of Gross Domestic Product (2019–2023 projected)

465

31.1 Overseas development assistance as proportion of GNI, selected countries 504 31.2 Composition of multilateral (black) and bilateral (striped) aid of Danish ODA

504

31.3 Danish ODA 2000–2017 with shifts in governments

514

31.4 Articles debating aid before and during election campaigns

516

32.1 Real Gross Domestic Product (GDP): Level and growth rates, 1966–2018

523

32.2 The employment and unemployment rate, 1966–2018

523

32.3 Price and wage inflation and interest rates, 1966–2018

524

32.4 Current account and public sector balance, 1966–2018, per cent of GDP

524

32.5 Recipients of social transfers, 1960–2017

536

34.1 The Danish flexicurity model

566

34.2 Public expenditure in labour market policy interventions by broad type in GDP for selected EU member states—in per cent

568

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

xviii   list of figures 36.1 Voters’ attitudes to whether government spends too much or too little money on the healthcare sector

604

37.1 Immigration to and emigration from Denmark, 1946–2018

610

37.2 External boundary: Immigration as a serious threat to ‘our distinct national character’, 1967/88–2015

616

37.3 Internal boundary: Immigrants’ equal access to social benefits, 1994–2015

616

39.1 Environmental taxes—per cent of GDP

646

39.2a Evolution of SO2 emissions

650

39.2b Evolution of NOX emissions

650

39.3 Macroalgae (total cover) and eelgrass (maximum depth limit) in marine waters

651

39.4 Share of fauna classes in water courses (per cent)

652

39.5 Denmark’s gross energy consumption and production of primary energy

658

39.6 Denmark’s production of renewable energy

660

40.1 Major Danish research policy reforms

669

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

List of Tables

3.1 Basic Asymmetries of the Community of the Realm 2017

33

5.1 The Overall Structure of the Danish Electoral System

59

5.2 The Development in the Main Features of the Electoral System, 1920–2020 60 5.3 Changes in Lijphart’s Four System Change Indicators and Average Values for Gallagher Disproportionality Scores, Effective Number of Parties (N(v) and N(s)), and r

61

5.4 The Official Result of the 2019 Folketinget Elections and the Computation of Compensatory Seats

66

5.5 Selected Features of Scandinavian Electoral Systems

72

6.1 Turnout across Specific Groups, 2017 Local and 2015 National Elections

81

6.2 Overview of the Experiment’s Total Effects

83

8.1 Government Formation in Denmark

110

8.2 Government Termination in Denmark

113

10.1 Functions of Central Government, Regions, and Municipalities

143

10.2 Gross Current and Capital Expenditures of Municipalities and Regions (2019 Budget)

151

10.3 Variations in Selected Municipal Expenditure and Taxation Levels (2019 Budget)

152

10.4 Municipal Revenue Sources (2019 Budget)

153

11.1 Number of Commissions, Committees, etc. with Interest Group Representatives, 1975–2015

165

12.1 Bureaucratic Reforms, 1660–1866

187

15.1 Party Unity in Denmark 2007/08–2017/18, Per Cent

232

15.2 Voting Patterns on Passed Bills in the Danish Parliament, 1973–2018, Per Cent

236

19.1 Positions of the Liberals and Select Other Danish Parties on Important Issue Dimensions, 2014

301

21.1 Voter Demographic and Voter Opinions of the Red-Green Alliance Compared to the Other Parties in the Danish Parliament

334

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

xx   list of tables 21.2 The Position on the EU, Redistribution, and the Environment and the Issue Importance for Six Parties in Denmark, Sweden, and Germany

336

24.1 The Results of the 2017 Local Elections by Political Party Aggregated for All Ninety-Eight Municipalities and Five Regions Compared to Vote Intention in National Election

386

24.2 Inter-Level Split-Ticket Voting at the Municipal and Regional Elections of November 2017 and a Fictive Election for the Folketinget on the Same Day

386

24.3 Inter-Level Split-Ticket Voting at the Municipal Elections of 2017 and a Fictive National Election (Vote Intention) Combined with the Voters’ Focus on Local Candidates and Issues or National Politics when Casting the Municipal Vote

388

24.4 Number of Lists Running and Getting Elected in the Ninety-Eight Local Elections on 21 November 2017

389

24.5 Preferential Voting for Male and Female Candidates at the 2017 Danish Local Elections

393

24.6 Mayoral Parties after the Municipal Elections, 1970–2017, Per cent

395

25.1 National Referenda in Denmark since 1915

402

25.2 Yes Vote by Party Preference

408

25.3 Voter Attitudes towards Different Issue Areas, 1992 and 1993

409

27.1 Overview of Group Types in 1975 and 2010

438

27.2 Importance of Contacts to Different Actors

442

27.3 Access to the Administration and the News Media, 1975 and 2010

445

30.1 Comparison of EU Policies

488

30.2 Characteristics of the Analysed EU Policies

499

31.1 Danish Development Strategies Since 2000

506

34.1 Trade Union Membership from 2000–2018 (in Thousands, ‘Share of ’ in per cent)

570

34.2 Private Sector Collective Agreement Coverage from 1998–2015

572

37.1 Percentage of Danish Population with Danish and Immigrant Origin, 1980–2018611 37.2 Requirements for Citizenship in the Scandinavian Countries, 1995–2019

614

37.3 ‘Danish Behaviour’ as a condition for Danish citizenship

615

37.4 Public Opinion in Selected Western European Countries, Percentage Restrictive/Anti-Immigrant Responses, 2014

617

40.1 Competing Conceptions of Research Quality and Societal Impact

667

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

chapter 1

W h y a Book on Da n ish Politics? Peter Munk Christiansen, Jørgen Elklit, and Peter Nedergaard

The purpose of this handbook is to take stock of what we in the social science community know about Danish politics in recent years. We have invited authors with a track record of fine publications in relevant areas on Danish politics in order to thoroughly and critically scrutinize a large number of aspects of Danish politics. In terms of dis­cip­lin­ary coverage, political science is obviously well represented among the authors. However, a project of this kind must also take into account contributions from other disciplines and fields, and we have invited contributors from economics, law, history, and educational science, as well as from disciplines close to mainstream political science such as political economy and political sociology. Why a handbook on Danish politics? On the one hand, Denmark shares important features with other smaller—Denmark has 5.8 million inhabitants—countries in NorthWestern Europe such as having a parliamentarian democracy, a PR electoral system, a multiparty system, a high level of public spending, a relatively stable party system, relatively strong interest groups, and a monarch with nothing but ceremonial functions. Denmark is also an EU and NATO member. Like most other countries in this group, Denmark runs a quite well-functioning welfare state, and the standard of living is high, as documented in all international comparisons. In particular, Denmark shares characteristics with the two other Scandinavian countries, Norway and Sweden, such as a relatively strong Social Democratic party, a generous and universal welfare state, a long and strong corporatist heritage, and high levels of general and political trust. On the other hand, Denmark also turns out to be different in some aspects, i.e. to deviate on some dimensions when compared to comparable countries like Norway, Sweden, Iceland, Finland, the Netherlands, Belgium, Austria, and Switzerland—and sometimes even Germany, the United Kingdom, New Zealand, and Canada.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

2   Christiansen, Elklit, and Nedergaard We shortly discuss some of these features below: a world record in minority g­ overnments and yet a stable political system; a party system that in some aspects deviate from that of similar countries; continued reservations towards the EU despite close to 50 years of EC/EU membership; a remarkable balance between tolerance and liberalism on the one hand and what is sometimes perceived as xenophobic policies on the other; a high degree of flexibility on the labour market as far as hiring and firing is concerned, which is even supported by both employers and trade unions; an activist foreign policy—including military means—taking place in both peace-keeping and active warfare; equality of the sexes through early female employment and generous day care on the one hand and lack of a proactive gender policy on the other; and finally, the largest local and regional government sector in the world. Generally, the authors have tried to cover both when they—through their scholarly glasses—see that something is rotten in the state of Denmark (William Shakespeare in Hamlet) and when Denmark is punching above its weight (Barack Obama during a visit from the Danish PM in 2011). Denmark has been characterized by extreme minority parliamentarism since the 1950s and actually holds a world record in minority governments. Denmark has not had a majority party in Parliament since 1909, and since 1960, it has only had majority (co­ali­tion) governments in 1968–71 and 1993–4. This implies that the political parties have developed ways of creating legislative majorities and avoiding parliamentary majorities willing to terminate the lives of the incumbent governments. Negative parliamentarism, i.e. that the government must not have a majority against it in parliament (but it does not need an explicit majority), is one instrument that has proved useful on many occasions. Another instrument is the use of settlements on policy issues between the incumbent government and different and changing sets of opposition parties. The number of parties in Parliament—oftentimes more than six, and no less that ten after the 2019 general election—which to some degree is caused by the electoral system being a system of proportional representation with a low effective threshold, does not create obstacles for the development of a well-functioning polity in the short term. However, long-term perspectives are sometimes lost, and it is not always easy to raise accountability issues in relation to individual political parties or governments’ policy outputs. Several of the chapters deal with these issues. At the June 2019 election, parties established before 1960 received 72.2 per cent of all votes. While this hints towards stability, the Danish party system has also shown remarkable changes. An anti-establishment, populist party, the Progress Party (Fremskridtspartiet), entered Parliament as early as 1973 with 15.9 per cent of the vote. It developed into an anti-immigrant party. In 2001, a splinter from the party, the more centre-oriented Danish People’s Party (Dansk Folkeparti), became a permanent support party for centre-right governments during 2001–11 and 2015–19, mainly on an anti-immigrant platform. While most European countries have experienced antiestablishment and anti-immigrant parties in recent years, Denmark deviates by experiencing this early on and through an early integration of the Danish People’s Party into mainstream politics. The Danish party system also deviates from most

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

Why a Book on Danish Politics?   3 other European countries by the absence of a Green party, but environmental issues were soon adopted by other parties. Denmark’s membership of the European Union also differs from that of other member states. Denmark became a member of the then European Community as of 1 January 1973 after a clear majority for entry at the referendum on 2 October 1972. EC membership opponents continued to voice their views on membership and EC/EU policies, and it is only more recently that a broad political acceptance of membership has prevailed. This is partly to be explained by the Danish opt-outs concerning the euro, home and justice affairs, and defence policy, which were formulated as a response to the 1992 rejection of the Maastricht Treaty in a popular referendum. The electorate has confirmed on two later occasions (2000 and 2015) that the opt-outs are to remain part and parcel of the Danish relation to the European Union. Danish citizens generally remain enthusiastic about EU membership as well as the opt-outs. Denmark is—or used to be—known as a tolerant and liberal country. Almost all Danish Jews were saved during the Second World War due to popular, civil society, and political agency. Denmark was the first country in the world to free literate pornography (1967) and pictorial pornography (1969), and Denmark was the first country to legally accept same-sex relations in registered partnerships (1989). After the turn of the century, however, Denmark has also been associated with anti-tolerance and anti-liberal policies, particularly towards immigrants. Does this express political responses to popu­lar opinion, or is it the result of other political dynamics? Denmark is in the world’s absolute top league when it comes to taxation, and it runs one of the biggest publicly financed service sectors in the world, which might be due to minority governments’ never-ending quests for support on legislative proposals. In the late 1970s and early 1980s, the Danish economy was at the ‘brink of the abyss’ as expressed by a former Social Democratic minister of finance. The economy recovered during the 1980s and 1990s due to fiscal austerity and promotion of flexibility on the Danish labour market. There may be many reasons to study Danish politics. We have mentioned some. We encourage readers to delve further into the book on their own to discover other findings in the various chapters.

Structure of the Handbook The Handbook of Danish Politics thus aims at providing overviews and analyses of a number of topics, patterns, and developments which together form what we term Danish politics, its background, and actual content. The thirty-nine substantial chapters, i.e. excluding this opening chapter and its concluding sister at the end of the volume, in the three sections all contain a reflection of state-of-the-art research on their respective topics, but they also present new arguments and insights, making contributions in their own right.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

4   Christiansen, Elklit, and Nedergaard The three sections are: I. Polity (Chapters  2–13): how is Denmark structured as a state (e.g. political ­institutions, constitutional order, and power division)? II. Politics (Chapters 14–27): what are the main elements of Danish politics (e.g. the role of political parties, the electorate, interest groups, and the media)? III. Policies (Chapters  28–40): what are the most distinctive policies in Denmark (e.g. social policy, economic policy, labour market policy, immigration policy, and foreign policy)? The section on polity contains chapters dealing with the political and administrative organization (or structuring) of the state. It thus provides a chapter on the Constitution (Chapter 2), and supplementing chapters on the special relationship between the three parts of the realm, Denmark, Greenland, and the Faroe Islands (Chapter 3), and the monarch (Chapter  4). The following two chapters deal with the electoral system (Chapter 5) and election turnout (Chapter 6). Then follow five chapters on important political institutions: the Parliament (Chapter 7); the government and the prime minister (Chapter 8); the central administration (Chapter 9); local and regional government (Chapter 10); and corporatism (Chapter 11). The last part of the polity section deals with corruption (Chapter 12) and the system for coordinating Danish politics with that of the EU (Chapter 13). The section on politics includes chapters on the key actors and key policy processes in Danish politics. The first chapter gives an overview of the Danish party system (Chapter 14) followed by three chapters on different aspects of Danish political life: a chapter on minority governments (Chapter 15) followed by chapters on the role of classes (Chapter 16) and voters’ public opinion in a longitudinal perspective (Chapter 17). The following four chapters are on four political parties to each of which we have decided to devote a full chapter: the Social Democratic Party (Chapter 18) and the Liberal Party (Chapter 19)—the biggest parties on each of the main political blocs—and the Danish People’s Party (Chapter 20) and the Red-Green Alliance (Chapter 21). Then follows a chapter on voting behaviour (Chapter 22) and one on the role of gender in Danish politics (Chapter 23). Local elections (Chapter 24) and referendums (Chapter 25) are treated in separate chapters. The last two chapters in this section are on two other types of actors with important political roles, namely the media (Chapter 26) and interest organizations (Chapter 27). The section on policies presents a series of chapters on important policy areas in Denmark. The first three chapters are devoted to Denmark’s international relations: defence policy (Chapter 28), foreign policy (Chapter 29), Danish policies towards the EU (Chapter 30), and development policy (Chapter 31). Economic policy (Chapter 32) is a core policy in any country. As a member of the Scandinavian welfare state group, a chapter on the universal Danish welfare state (Chapter 33) is also a must. Labour market policy (Chapter 34) also shares traits with the other Scandinavian countries with its reliance on labour market flexibility as well as social security (i.e. flexicurity). The following

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

Why a Book on Danish Politics?   5 two chapters deal with two core policy areas of the universal welfare state: education (Chapter 35) and health policies (Chapter 36). The remaining chapters in this section deal with different sectoral policies: immigration policy (Chapter 37), agricultural and fisheries policy (Chapter 38), policies on the environment, energy, and climate (Chapter 39), and research policy (Chapter 40). The chapters in the three sections aim at reflecting the state-of-the-art research of the many different topics at the same time as they make new contributions in their own right, i.e. have a substantial argument and suggest future lines of research. As not everything can be covered in each chapter, readers might appreciate the cross-references between chapters. The volume concludes with a chapter (41) where the editors try to assess critically what the state of Denmark is at this point in time after the general elections of June 2019 and at the beginning of the third decade of the twenty-first century.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

pa rt I

P OL I T Y Section Editor: Jørgen Elklit

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

chapter 2

The Constit u tion Jens Peter Christensen

The Constitutional Act as the Framework for Danish Democracy The constitutions of western democracies often begin with the mention of the fundamental principles on which the country’s form of government relies. The Swedish Constitution, for instance, starts out by stating that all public authority stems from the people and that Swedish democracy is based on the free formation of public opinion and the common and equal right to vote. The second paragraph of the Norwegian Constitution states that the purpose of the Constitution is to safeguard democracy, the constitutional state, and human rights. The Constitutional Act of Denmark might well have had a similar introductory passage, yet it does not. The Danish Constitutional Act has no such distinct paragraph about the fundamental principles for the country’s form of government. Section 2 of the Danish Constitutional Act simply notes that ‘the form of government is limited monarchy’. Nowhere in the Danish Constitutional Act does the word ‘democracy’ appear. One reason for this is that many of the constitutional provisions in the current Constitutional Act appear almost entirely unchanged from when they were first written as part of Denmark’s original Constitutional Act of 1849. During subsequent amendments, the most recent of which is the constitutional amendment of 1953, the choice has been to elaborate on the existing wording of the Constitutional Act without any comprehensive changes to the language. The Constitutional Act is the result of a range of political compromises and reflects a number of historical strata in the constitutional evolution. For that reason, the Constitutional Act does not represent an unambiguous set of values or any singular ideol­ogy. However, we can still say that the Constitutional Act rests on at least two fundamental principles that serve as the backbone of Danish democracy and the Danish constitutional state.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

10   Jens Peter Christensen The first principle is that power derives from the people and that the power vested in authorities must be exercised within the limits of the law. This principle finds expression in the Constitutional Act’s provisions about how voters elect members of the Folketinget (the Danish Parliament) in general, secret, and direct elections. Together with the government, the Folketinget holds legislative power and can, by virtue of the constitutional rule about parliamentarism, at any point express its lack of confidence in the government, resulting in the government stepping down or calling for a new parliamentary election. Independent courts are responsible for ensuring that power is exercised within the bounds of the law and the Constitution. The second fundamental principle is that citizens are guaranteed a certain minimum number of rights. In particular, the Constitution points to two types of rights. One type refers to political rights: freedom of speech, freedom of assembly, and freedom of association. The primary purpose of these rights is to ensure the proper functioning of democracy since the constitutional rules about parliamentary elections would be rendered rather useless without citizens’ right to discuss political issues, form political parties and organizations, and assemble in groups. The second type of rights pertains to rules about personal freedoms and to the inviolability of property and the home. These rights are primarily designed to protect the individual against acts of random intervention by the state. The Constitution’s provisions are an expression of the fact that certain important and fundamental societal questions have been legally regulated in a way that renders them out of the realm of the common majority’s political consideration and decision-making. In this way, the Constitution serves as a point of intersection between law and politics. The Constitution provides a framework for constitutional life, but it is important to note that the Constitution is only a framework and not the entire picture. Below is a brief introduction to the provisions in the Danish Constitution about the highest state authorities and their competence as well as its provisions about constitutional rights, with continuous attention to the social and political reality to which these provisions pertain. The chapter builds in large part on Christensen et al. (2016). The Constitutional Act can be found in English with accompanying explanations at www. FT.dk (Folketinget n.d.).

The History of the Constitutional Act Denmark’s current Constitution dates back to 1953. As already mentioned, many of the clauses in the current Constitutional Act remain virtually unchanged from Denmark’s original Constitutional Act of 1849. That original constitution represented a break with the constitution of absolute monarchy, the King’s Law (Lex Regia) of 1665. The Constitution of 1849 introduced a strangely broad and democratically attuned form of government for the time with its introduction of common voting rights in parliamentary

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   11 elections, which consisted of two chambers at the time, namely the Folketinget and the Landstinget. However, in accordance with the period’s existing customs and beliefs, this right was limited to reputable Danish men above the age of 30 with the exception of servants without their own household, recipients of social welfare, and those lacking legal capacity. While all power during the absolute monarchy had unambiguously been in the hands of the king, Section 2 of the Constitutional Act of 1849 introduced a tripartition of power. This meant that the legislative power was shared between the king (the government) and the parliament, the executive power belonged to the king (government), and the judicial power resided with the courts. This clause can also be found in the current Constitutional Act under Section 3. Aside from a number of amendments during the first 15 years after the Constitution of 1849 was written (as a result of specific issues pertaining to the duchies of SchleswigHolstein and Lauenborg occasioned by Denmark’s loss in the 1864 war against Prussia and Austria), the Constitution has only been amended four times since 1849, namely in 1866, 1915, 1920, and 1953. The amendment of 1866 was necessitated by Denmark’s defeat to Prussia, which led to the loss of the above-mentioned duchies. This reduced the kingdom’s territory by 40 per cent and led to the loss of approximately one million of its three million citizens. Moreover, the 1866 amendment was sustained by a sense that the Constitutional Act of 1849 had gone too far in the direction of a democratically conceived constitution. The amendment of 1866 led to the abolition of equal voting rights for the Upper House, Landstinget. It introduced privileged voting rights so that voters with substantial incomes or high tax payments were granted a greater level of influence on the distribution of seats in Parliament, particularly in rural districts where those individuals served directly as electors. At the same time, the king (the government) was granted the authority to appoint 12 of the Landstinget’s 66 members. This laid the groundwork for the constitutional battle between the Folketinget and Landstinget which lasted from 1872 until the introduction of parliamentarism as political practice in 1901 when the party holding the majority in Parliament (the Liberals) formed a new government. The constitutional amendment of 1915 broke with the 1866 Constitution’s privileged voting rights status for the wealthy and a rejection of the system whereby the king (government) appointed a certain number of members of the Landstinget. Furthermore, it afforded voting rights to women. In an effort to inhibit comprehensive future amendments, the 1915 Constitution introduced a rule that makes it very difficult to amend the Constitutional Act. The rule can be found in Section 88 of the current 1953 Constitution, according to which constitutional amendments require approval in the Folketinget and a subsequent parliamentary election. The proposed amendment has to then be approved without any changes by the new Folketinget. Finally, it has to be put up for a referendum and be approved by a 40 per cent majority of all voters (prior to 1953, this was 45 per cent). As part of the agreement of European peace at the conclusion of the First World War , the minor constitutional amendment of 1920 was necessitated by the desire to regain control of the southern part of the country bordering Germany as a constitutionally

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

12   Jens Peter Christensen integrated part of the Kingdom of Denmark following special referendums about where to draw the border. The current Constitutional Act of 1953 barely survived the challenging amendment process. The reason that it eventually found approval had to do with the fact that it was lumped in with an amendment to the Act of Succession. In accordance with the Constitution of 1920, the law of succession only pertained to men. However, King Frederik IX, who became king in 1947, only had daughters, and in order for his eldest daughter Princess Margrethe to inherit the throne, the Act of Succession would have to be amended. Because such an amendment was very popular among the public, politicians chose to pair it with the constitutional amendment, presuming the amendment to the Act of Succession would be so popular among the Danes that it would secure approval for the entire constitutional amendment along with it. This is exactly what happened. The proposed amendment to the Constitution was approved in the referendum with the needed support, but only barely. Despite the popular amendment to the Act of Succession, the proposed amendment only found support by 45.76 per cent of registered voters. If 19,000 of those who voted in favour had stayed home, the constitutional amendment would have been rejected. Only 12.3 per cent of voters voted against the amendment. Other significant constitutional amendments to the 1953 Constitution were the transition from the two-chamber system to a unicameral system, a lowering of the voting age from 25 to 23, and the constitutional consolidation of parliamentarism. Most im­port­ant, perhaps, was the provision to make it possible to delegate sovereignty (in the form of legislative, executive, and judicial power) to international authorities without a constitutional amendment (Section 20). That change has been hugely significant with regard to Denmark joining the EC in 1972 (now the EU), as well as Denmark’s support for subsequent changes to the EU treaty. The constitutional amendment of 1953 resulted in the following wording of Section 1 (Folketinget n.d.: 2): ‘This Constitutional Act shall apply to all parts of the Kingdom of Denmark.’ This means that the Constitution applies to Denmark, the Faroe Islands, and Greenland. Within the scope of the Constitution, the Faroe Islands and Greenland have their own individual home-rule agreements. Constitutionally, these agreements have the status of laws, but their political status makes it inconceivable that they could be repealed or amended in any substantial way by the Danish legislative body without the participation and support of the Faroe Islands and Greenland.

The King and the Monarchy As already mentioned, Section 2 of the present 1953 Constitution (Folketinget n.d.: 2) states: ‘The form of government shall be that of a constitutional monarchy.’ The wording hints at the autocratic rule before 1849 and confirms that while Denmark is a monarchy whose monarch is head of state (currently Queen Margrethe II), the monarch’s authority

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   13 is subject to the limitations inherent in the constitutional system established by the other constitutional provisions. These limitations are far-reaching and mean that the constitutional role of the monarch is merely symbolic and ceremonial. The monarch is not granted any competence to act in state affairs independently of the government and ministers. This arrangement is not a result of a new interpretation of the Constitution. The legal literature had already established 150 years ago that independent competence for the monarch would require ‘an abandonment of the entire constitutional system’. However, in practical terms, for several decades after 1849, the transition away from absolute monarchy was characterized by the king’s ability to influence the government and ministers. The monarch’s lack of competence to act independently of the government and ministers in state affairs is laid out in Sections 12, 13, and 14 of the Constitution. The central tenet of these sections is that the ministers alone are liable and that the monarch is exempt from liability. The monarch’s signature is required for a number of decisions regarding legislation and administrative decisions, as well as for certain decisions made in regard to particular constitutional provisions such as the appointment and dismissal of ministers (Section 14), decisions about foreign policy (Section 19), and the decision to call a general election (Section 32, Subsection 2). However, the monarch does not have the freedom to withhold his or her signature, and thus, is not entitled to veto decisions. If a monarch refuses to follow a recommendation introduced by a minister, the monarch is constitutionally obligated to comply. If the monarch is unwilling to do so, he or she is forced to abdicate. The monarch’s private dispositions fall outside the rules about the monarch’s signature being accompanied by that of a minister. The grey area between private dispositions and acts of the state contains decisions about the appointment of royal court officials and awarding royal orders and decorations. In accordance with tradition, these decisions are made without the countersignature of ministers. In addition to the aforementioned decisions regarding legislation, etc., the monarch’s participation in state affairs primarily takes the form of very concrete actions. An ex­ample would be the monarch’s official visit to other nations and official state visits in Denmark, participation in a number of official events, and delivering speeches at historical commemoration ceremonies or the annual New Year’s speech on national television. In such instances, the principle about the monarch’s exemption from liability and the liability of ministers requires the monarch to secure a minister’s approval, typically from the prime minister. However, in cases when the statements are not political in nature, or in the case of representative functions not perceived to carry any political judgment, the monarch typically acts without the prior approval of a minister. Overall, the general rule is that the monarch should be kept free of political involvement even when acting in the context of official state affairs. As such, the government should not misappropriate the monarch for the purposes of delivering partisan political viewpoints. In practical terms, the monarch stays away from politically controversial subjects by referencing only what falls within broad political consensus. The monarch is meant to represent a uniting, not a dividing, role in state affairs. It is generally agreed upon that Denmark’s current monarch Queen Margrethe II has mastered this task.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

14   Jens Peter Christensen Queen Margrethe II took the throne in January 1972 following the death of her father King Frederik IX. The accession of the new queen was a result of the amendment to the Act of Succession included in the constitutional amendment of 1953 which introduced female royal succession. A subsequent amendment to the Act of Succession in 2009 introduced full equality between men and women. Queen Margrethe II’s oldest son Crown Prince Frederik is next in line for the throne.

The Folketinget In his textbook on constitutional law from 1959, Danish Professor of Constitutional Law Alf Ross wrote that the Folketinget is ‘the central wheelhouse of the state machine’ (Ross 1959: 216, own translation). Political scientists know that a lot more can be said about this point. In any case, it remains a fact that the Folketinget is given a prominent position in the Danish Constitution. By virtue of parliamentarism, the Folketinget is in charge of the government. The Folketinget has the power to dismiss the government or its individual ministers at any given time. And because of Section 6 in the Constitution, the Folketinget shares legislative power with the government. As the only constitutionally established state body, the Folketinget represents a direct popular mandate, and on that basis, it can be said to be the most significant democratic function. The Folketinget consists of a 179-member assembly. Members are elected for a period of four years although the prime minister can call a general election at any point, leading to the annulment of the parliamentary mandates once the election takes place. Section 29 of the Constitution outlines a number of preconditions for an individual’s eligibility in relation to the Folketinget. Positive requirements include Danish citizenship, permanent domicile in the country, and being of voting age. Negative requirements include not having been declared incapable of conducting his or her own affairs (i.e. having lost legal capacity) and not having lost the right to vote due to prior convictions or the receiving of welfare benefits. However, there is no longer any law pertaining to these two final requirements. The legal voting age is not directly established in the Constitution but is based on the Parliamentary Election Act. In accordance with Section 29 of the Constitution, changes to the legal voting age require not only that this change be signed into law but also that this law is sent to a referendum. In 1953 when the Constitutional Act was first written, the voting age was 23. Today, it is 18. Anyone who is eligible to vote in parliamentary elections is also electable to the Folketinget unless the person has been convicted of an act that makes him unworthy to be a member of the Folketinget in the eyes of the public, cf. Section 30 of the Constitutional Act. In accordance with Section 33 of the Constitutional Act, the Folketinget singlehandedly determines questions of electability. Since 1953, there have been five instances when the Folketinget has denied electability status to one of its members.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   15 The Constitutional Act’s provisions about elections to the Folketinget can be found in Section 31, whereas the rules are established in more detail in the Parliamentary Election Act. According to the Constitution, elections to the Folketinget must be ‘general, direct, and secret’ (Folketinget  n.d.: 16). The term ‘general’ implies that the authorities are expected to conduct elections in such a way that the right to vote will be available to the greatest possible number of citizens. ‘Direct’ elections means that the votes cast by citizens directly decide which parties and candidates are elected to the Folketinget. ‘Secret’ elections are defined in opposition to the open voting practiced before 1901; it entails that voters cast their ballots in a closed voting booth and that their ballot cannot subsequently be identified. Section 31 outlines the general requirements of the electoral system to ensure that the views of voters—which in practical terms mean the parties—are given equal representation in the Folketinget. The Constitution does not pinpoint a particular model of proportional representation. This is established in the Parliamentary Election Act. The electoral system combines 135 constituency seats and 40 compensatory seats. The latter are distributed among parties whose constituency seats have not granted them the number of seats to which they are entitled under proportional representation. The Parliamentary Election Act contains an electoral threshold which means that as a general rule, parties with less than 2 per cent of the total vote are unable to obtain compensatory seats. The purpose of this threshold is to prevent unnecessary division of the Folketinget into small parties that might complicate the formation of sustainable governments and effective legislation (see also Elklit 2020). The Constitutional Act does not make explicit reference to political parties, but in practical terms, the work of the Folketinget is organized around party groups. According to Section 56, members of the Folketinget are bound solely by their own consciences, meaning that members are not legally obligated to adhere to the party line when it comes to voting in Parliament. However, there tends to be a high level of party dis­cip­ line, and members of the different party groups most often vote in unison. Party dis­cip­ line is further solidified by the fact that it is extremely difficult to become elected to the Folketinget as an independent candidate. It has only happened once during the current 1953 Constitution.

The Legislative Process The normal legislative process is only regulated by a very few constitutional provisions. Any member of the Folketinget is entitled to introduce a bill (Section 41), as are the government’s ministers (Section 21). In reality, most bills are introduced by ministers. Once a bill has been introduced to Parliament, the Constitutional Act contains only one single provision about the Folketinget’s normal handling of the bill, which is that it has to be read three times in Parliament before it can be passed. The Standing Orders of the Folketinget, on the other hand, contain detailed rules on how to deal with a bill proposal,

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

16   Jens Peter Christensen including its assignment for study to a committee. Section 48 of the Constitutional Act establishes that the Standing Orders are decided by the Folketinget, but the Constitution does not specify any requirements about the content of the Standing Orders. With respect to the passing of bills, the Constitution requires that more than half of the 179 members of the Folketinget be present and take part in voting (Section 50). In order for a bill to become law once it has been passed, a minister and the monarch must ratify it with their signatures. In certain cases, the normal legislative process can or must be dispensed with in accordance with particular constitutional provisions, granting the parliamentary minority the opportunity to have their views considered. First, the Constitution in Section 41 gives 2/5 of the members of the Folketinget the right to demand that the third reading of a bill be postponed until at least twelve working days after the bill’s second reading. The purpose of this is to allow for a more thorough reading and to create public debate about the bill’s content and fate. This provision has been employed quite a few times. Second, according to Section 42, a minority of 1/3 of members of the Folketinget (60 MPs) can demand a referendum on a bill that has been passed. This provision was added to the Constitution in 1953. The most exciting aspect of this provision is that it has only been used once, in 1963, when the conservative opposition requested that four bills concerning the public regulation of land ownership were submitted for a referendum. A large majority of the voters rejected all four bills. Despite this success, no other parliamentary minority has made use of the provision since then. A deceased former Danish foreign minister—and a big proponent of referendums—explained this by arguing that a parliamentary minority loses one of its trump cards in the next election campaign if it has already pushed through its policies in a referendum. In some case, the mere threat of a referendum might also do the trick. Third, it is required that the normal legislative process be dispensed with when constitutional amendments are introduced (Section 88), when delegating sovereignty to international organizations (Section 20), and in proposed amendments to the voting age (Section 29). In each of these cases, the Constitution calls for a referendum, though only in cases about delegating sovereignty if the bill does not find support from a 5/6 parliamentary majority. As for the fourth and final provision, the Constitution contains a special minority clause concerning laws about expropriation (Section 73). In such cases, one third of the members of the Folketinget can request the postponement of the ratification of a passed bill until after a parliamentary election and the bill’s subsequent passing by the new parliament. This provision has been used a few times. In addition to the aforementioned provisions about referendums, the Constitution allows a parliamentary majority to call a consultative referendum. This option was used in 1986 when a consultative referendum was held about Denmark’s support for the Single European Act which introduced the European single market. The government was in support of Denmark’s inclusion, but a majority in Parliament was opposed. The voters said yes and the parliamentary majority complied with the voting majority.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   17 Thus,  the minority government managed to successfully use the referendum as a weapon to defeat the parliamentary majority.

Constitutional Provisions on Parliamentarism The structure outlined in the Constitutional Act for political parliamentary activities is simple, and it is defined by two unambiguous constitutional provisions. First is Section 15, Subsection 2, which establishes the Parliament’s right to make a motion of no confidence against the prime minister at any time. According to the Constitution, such a motion means that the government must step down or that the prime minister must call for a parliamentary election. The second provision is Section 32, Subsection 2, which establishes the prime minister’s authority to call for a parliamentary election at any time. These very simple provisions do not leave any doubt as to where to draw the line between law and politics. This in itself is an important quality about provisions that have been designed to regulate a subject as complicated and conflict-riddled as the col­lab­ora­ tion between government and parliament. These provisions have managed to create a framework for highly changeable parliamentary situations. This became particularly clear in the 1980s when Conservative Prime Minister Poul Schlüter headed a number of minority governments. Back in 1982, it was still possible (and rightfully so) to state in a political science textbook that a government would never accept defeat in a vote on a significant issue. ‘Thus, we can say,’ the author writes, ‘that any vote held in Denmark represents a vote of confidence’ (Worre 1982: 102, own translation). However, the ink was barely dry when parliamentary developments put the claim to shame. On a great number of issues, primarily relating to national security policies, the government was in the minority in Parliament without it resulting in the government stepping down. The same thing happened in votes on energy, environmental, and legal policies. During the period 1982–88, the government lost 1 in 12 votes in Parliament (Damgaard 1990: 28). In total, in the ten years 1982–92, the government lost more than 100 votes. This period in the 1980s has been described as a period of ‘reverse parliamentarianism’ (Christensen 1993: 4). The term refers to the fact that the government almost played the role of the opposition on a number of issues, whereas the opposition played the role of government. An extended period of reverse parliamentarism of this kind has not existed since then, though there have continued to be instances where the government’s proposals have been voted down without it resulting in a change of government. There are different views on the expedience of this kind of parliamentarism, but in terms of constitutional law, there is no disagreement. If it is not possible to create a parliamentary majority in favour of a vote of no confidence, then the government stays and accepts, as it did in the 1980s, in order to figure out how to administer the policies of the opposition.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

18   Jens Peter Christensen The Constitutional Act’s Section 15, Subsection 2, establishing the government’s obligation to step down or the prime minister’s obligation to call an election if the Folketinget wins a vote of no confidence is supplemented by Section 15, Subsection 1, which allows the Folketinget to adopt a vote of no confidence against an individual minister. In such cases, the minister must resign. The Danish Folketinget has never made such a motion against a minister. This does not mean that this aspect of the political ministerial accountability is not effective. In fact, we might say that what happens in practice is that a minister makes sure to resign ‘voluntarily’ before he or she is forced to do so. Thus, this constitutional provision proves effective by virtue of its very existence. Since the introduction of parliamentarism in 1901, the Folketinget has only passed a vote of no confidence against a prime minister on three occasions. The first time was in 1909, the second time in 1947, and the third and most recent as far back as 1975. However, a government might run into enough resistance that it gives up without a fight and without calling an election. This happened when Social Democratic Prime Minister Anker Jørgensen stepped down in 1982 and handed over power to Conservative Poul Schlüter. Alternatively, the government might resign because it risks otherwise being served a vote of no confidence, thus pre-empting the lack of confidence. This was part of the reason that Prime Minister Poul Schlüter stepped down in 1993 following the publication of a highly critical report about a wide-ranging political scandal (the Tamil Case).

Forming a Government Much like the constitutional provisions that concern parliamentarism, the provisions dealing with the process of forming a government are exceedingly simple. The framework for forming a government is made up by Sections 14 and 15 of the Constitutional Act. Section 14 states that the king (i.e. the prime minister) is responsible for appointing the prime minister and other ministers, while Section 15 concerns parliamentarism and entails that no government can be appointed that is presumed to be opposed by the majority in the Folketinget. In cases when a government calls an election and subsequently wins the election, backed by the same parliamentary majority as before the election, the government is allowed to continue uninterrupted. The situation gets more complicated if the government loses the majority in the election, and there is no obvious majority for a new government—a scenario that has occurred frequently. In such cases, party leaders begin negotiations to determine the different possibilities for forming a government. Over the course of the past 100 years, it has become a tradition for party leaders to inform the monarch—currently Queen Margrethe II—of their choice for prime minister. This is referred to as ‘The Queen’s Round of Consultations’ (Christensen 2017: 34). However, this term is a bit misleading. The party leaders’ negotiations do not take place in the presence of the queen. The entire procedure of forming a government is the responsibility of the sitting prime minister who makes all decisions about the process.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   19 Political and strategic games have often been a characteristic feature of the process of forming a government. However, the legal basis for forming a government is straightforward and can be summed up in three simple constitutional rules of constitutional law: 1) The entire procedure of forming a government is the responsibility of the sitting prime minister. 2) No one who is presumed to be met by a vote of no confidence in Parliament may be nominated as a candidate for prime minister. 3) The new prime minister nominates himself (or herself), and thus, is responsible for ensuring that he or she is not opposed by a majority in Parliament. During the process of forming a government, the simplicity of these provisions do not prevent frequent and competing claims in the press and in the political debate about the existence of other constitutional provisions. These claims, however, are not valid. Like elsewhere, the Constitution only provides a framework for the political situation, not the entire picture.

Parliamentary Oversight of Government and Ministers The constitutional provisions in Section 15 regarding parliamentarism according to which a majority in the Folketinget can declare their lack of confidence in both individual ministers as well as the prime minister establish the Folketinget’s ultimate authority and oversight over the government. As already mentioned, this weapon is generally not employed in practice but is effective by virtue of its very existence. On a day-to-day basis, parliamentary oversight of the government plays out through questions addressed to the ministers, consultations with standing committees, and parliamentary interpellations. Only the latter form of oversight is constitutionally regulated as Section 53 states that any member of Parliament, with the consent of Parliament, may submit for discussion any matter of public interest and request a statement thereon from the relevant minister who is obligated to answer truthfully. In accordance with the Standing Orders of the Folketinget in Section 20, any member of Parliament can address questions to the ministers, and though the ministers are not legally obligated to respond, they generally do. Ministers are also obligated to respond truthfully. In accordance with similar provisions, the committee of the Folketinget can address questions to the ministers, and many committee questions lead to consultations with the ministers, allowing the committee the opportunity to use further questioning to seek a more in-depth explanation. There has been a significant increase in the Folketinget’s use of these oversight mech­ an­isms. Since the mid-2000s, the number of annual requests has been around 40–50.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

20   Jens Peter Christensen From 2015 to 2016, the number of Section 20 questions from individual members of Parliament was approximately 1,500, and the number of committee-generated questions in the past decade has oscillated between 9,000 and 16,000 a year (White Paper 1571/2018: 66–7). According to Section 15 in the Constitutional Act regarding parliamentarism, ministers are politically accountable to the Folketinget. This entails that the Folketinget, as the ultimate authority, can express its lack of confidence in a minister at any point. As a less drastic and more relevant response, the Folketinget, or a parliamentary committee, can advance different degrees of criticism of the minister. This criticism may be raised in the context of an upcoming consultation if the committee believes that the minister has not provided satisfactory answers, or if the committee believes that questions have surfaced about the minister’s administration of his or her duties that is deserving of criticism. In advancing such criticism or in the context of passing an actual vote of no confidence, the Folketinget is not legally bound by any rules except the majority rule. Thus, the minister does not have to have taken any legally reproachable action and dis­agree­ ment with his or her policies or discomfort about his or her personality is sufficient grounds. In terms of constitutional law, the political responsibility of ministers is not defined in legal terms. Legally, however, ministers are responsible for the conduct of the government as defined in Section 13 of the Constitution, which states that this responsibility is outlined in more detail by the law. Section 16 of the Constitution establishes that both the government and the Folketinget can indict ministers in cases of maladministration of office. Historically, the Folketinget has always been in charge of the indictment of ministers. Individual ministerial legal responsibility is further determined in the Responsibilities of Ministers Act of 1964. The central provision of the law is Section 5 which establishes when a minister can be criminally charged in the context of administering his or her office. Subsection 1 of the provision states that a minister will be criminally charged if they deliberately or by gross negligence abandon the duties placed upon them by the Constitution, the law, and their office. Moreover, Section 5, Subsection 2, establishes that a minister can be penalized if he or she provides false or misleading information to the Folketinget or fails to disclose information during a parliamentary proceeding that would be of significance to Parliament’s judgment of the case. In accordance with the Constitutional Act’s Section 16, criminal proceedings against ministers are decided by the High Court of the Realm. Section 59 of the Constitutional Act establishes that the High Court of the Realm consists of 15 Supreme Court justices and 15 members elected by Parliament from outside the Parliament. Since 1849, only five cases have come before the High Court of the Realm. The last two cases are from 1910 and 1995. In the most recent case—the so-called ‘Tamil Case’ regarding the Justice Department’s illegal order not to process a number of Tamil family reunification cases—the Justice Minister, who was also a former speaker, was sentenced to four months in prison. However, with age and health issues taken into consideration, the conviction was suspended.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   21 The law about a minister’s legal responsibility functions—not unlike the vote of no confidence concerning a minister’s political responsibility—more by virtue of its existence than through practical application. Additionally, the reality of the legal ministerial responsibility is that it will often add extra liability to a minister’s case if he is said to have broken the law and not only acted politically objectionably. In recent decades, this form of legally based criticism of ministers has frequently resulted in wide-ranging investigations carried out by an Inquiry Commission under the chairmanship of a judge. In a number of cases, the final report from such inquiry commissions has served as the basis for the Parliament, or subsections of Parliament, to politically criticize the involved minister. In the Tamil Case, this kind of report led to the decision by a majority in Parliament to put the former justice minister before the High Court of the Realm. The widespread use of inquiry commissions led by judges is a testament to the fact that the political responsibility of ministers is often framed in legal terms, meaning that the Parliament’s political criticism of a minister is significantly bolstered by arguments that are legal in nature. This is not all that strange since political criticism of a minister will appear more valid to the public if not merely advanced on the basis of political dis­agree­ment but also on the basis of unlawful actions by the minister in question.

Denmark’s International Relations and Delegation of Sovereignty The king ‘wages war and declares peace’. This was the brief and terse formulation of Denmark’s relationship to the world around it in the Constitutional Act of 1849, Section 23. Today, this relationship is vastly more complicated and so is the Constitution. The Constitutional Act contains two provisions about Denmark’s international affairs. The first one (Section 19) primarily addresses international cooperation in the form of international agreements, or so-called treaties. Denmark has entered into hundreds of such treaties, spanning everything from trivialities to some of the most significant agreements regarding Denmark’s position in the world, for example, treaties about Danish membership of the UN and NATO. The second provision (Section 20) pertains to a particular kind of treaty, namely treaties in which Denmark delegates so-called sovereignty to an international entity. Section 20 does not use the word sovereignty but speaks of ‘powers vested in the authorities of the Realm under this Constitutional Act’ (Folketinget n.d.: 9). Such powers refer mainly to legislative, executive, and judicial authorities. The most significant example of this kind of treaty is the one that establishes Denmark’s membership of the EU. The distinction between the two different kinds of treaties is extremely important since the Constitutional Act makes it very easy to enter into Section 19 treaties and very difficult to enter into Section 20 treaties.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

22   Jens Peter Christensen According to Section 19 of the Constitution, it is the responsibility of the government to conduct foreign policy. However, the Folketinget keeps the government on a short leash since any treaty of real significance requires parliamentary approval in the form of a common majority vote. That leash has become shorter over time, both legally and politically, as evidenced by Section 19, Subsection 3, of the Constitution and its provision about pre-emptive parliamentary oversight of the government’s foreign policy. The government is required to consult with the Foreign Policy Committee ahead of any major decisions on foreign policy. The Committee is elected by parliament and consists of 17 members of Parliament. The provision about the Foreign Policy Committee was added to the Constitutional Act in 1953. The government’s obligation to ensure advance consultation with the Foreign Policy Committee is a legal obligation, meaning that if the government fails to do so, it could, in principle, lead to impeachment proceedings and criminal liability. On the contrary, the government is not legally obligated to follow the advice of the Foreign Policy Committee. However, an intelligent government will, at least if it wants to remain in power. The provision from the old 1849 Constitution about the king’s ability to ‘wage war and declare peace’ has also been curbed. Section 19, Subsection 2, of the Constitutional Act of 1953 establishes that the government cannot use military force against any foreign state without the consent of the Folketinget. In March 2010 in a case about the Iraq war (U 2010.1547 H), the Supreme Court ruled that it is the government and the Folketinget who make decisions about the use of military force in accordance with the Constitution, and that international law, including the UN treaty, does not constitutionally limit that authority. As already mentioned, Section 20 of the Constitution concerns the specific type of treaties in which Denmark delegates legislative, executive, or judicial power to an international entity such as in the case of Denmark’s membership of the EU. This provision was added to the Constitution in 1953 with the aim of making it easier to join this kind of supranational collaboration. Without Section 20, this would require a constitutional amendment. Despite the fact that Section 20 has made this process less complicated, it is still quite challenging to enter into Section 20 treaties. According to Section 20, Subsection 2, a 5/6 parliamentary majority—i.e. at least 150 members of Parliament—are required to vote in favour of a bill that surrenders authority to an international entity. If the bill only receives a common parliamentary majority, it must be submitted to the electorate in a referendum. The bill is rejected if a majority of the electorate, consisting of a minimum of 30 per cent of eligible voters, votes no. Danish voters have often expressed significant scepticism about the idea of expanding EU collaboration. In 1992 in a referendum on the Maastricht Treaty, a majority of 41.7 per cent of eligible voters voted against the bill, while 40.5 per cent voted in favour. In 2000, in a referendum on Denmark’s participation in the euro monetary system, a majority of 46.1 per cent of eligible voters voted against the bill, while 40.5 per cent voted in favour. Similarly, in a referendum in 2015 on whether to maintain Denmark’s opt-out

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   23 concerning Justice and Home Affairs or replace it with an opt-in model, a majority of  37.5 per cent voted against any change, while only 33.1 per cent voted yes to the changes. At other times, the majority of voters have voted yes such as in 1972 when Denmark voted to become a member of EU, in 1993 in a referendum on the Maastricht Treaty supplemented by the so-called Edinburgh Agreement, and again in 1998 when the Danes voted in favour of the Amsterdam Treaty. A referendum in 2014 on Denmark’s participation in the Unified Patent Court also generated a majority of yes votes. Section 20, Subsection 1, of the Constitution limits the delegation of power to an international organization to ‘such an extent as shall be provided by statute’. In the Maastricht Treaty case from 1998 (U 1998.800 H), the Supreme Court ruled that the words ‘as shall be provided by statute’ are to be interpreted to mean that a positive de­limi­ta­tion must be made of the powers delegated, partly as regards the fields of responsibility and partly as regards the nature of the powers. Delimitation must enable an assessment to be made of the extent of the delegation of sovereignty. In other words, the government cannot simply write a blank cheque to the international authorities. On the other hand, according to the Supreme Court, the formulation ‘as shall be provided by statute’ cannot be interpreted to mean that the powers vested in the Danish state ‘can only be delegated to an international organization to a limited (lesser) degree’. However, the Supreme Court notes that it must be considered to be assumed in the Constitution that no transfer of powers can take place to such an extent that ‘Denmark can no longer be considered an independent state’. The question of where to draw the legal line for when ‘Denmark can no longer be considered an independent state’ is not an easy matter and quickly devolves into a political question rather than a legal one. Thus, the Supreme Court’s ruling established that determining the line has to ‘primarily be based on political considerations’. The reality is that the Constitution’s protections against wide-ranging surrender of authority rely primarily on the requirements of having a 5/6 majority in Parliament and having to call a referendum if a bill can only gain a regular parliamentary majority.

The Courts The provision in the Danish Constitutional Act about the separation of powers (Section 3) states that the Folketinget and the government share legislative power, the government holds executive power, and the courts have judicial power. Together with a number of specific constitutional provisions about the courts, this provision marks the fact that Denmark is not only a democracy but also a constitutional state. Traditionally, the central core of the concept of judicial power is the ruling in criminal cases and in legal disputes between citizens. In addition to this are cases about the legality of administrative decisions and cases concerning the constitutionality of legislation. Unlike many continental European countries, Denmark does not

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

24   Jens Peter Christensen have special administrative courts or a special constitutional court. These types of cases are handled by the regular courts. One word can be used to sum up the responsibility placed on the courts by the Constitution: independence. The Constitution specifically ensures this independence through a provision in Section 64 about how judges shall be governed solely by the law and cannot accept orders from the executive power. This provision also guarantees the judge’s personal impartiality since a judge—according to the principal rule—can neither be dismissed nor transferred. Unless it is part of a more comprehensive reorganization of the courts, a judge can only be dismissed through a court ruling. The Constitution does not mention the process of nominating judges. In 1999, a bill was passed that established a Judicial Appointments Council, which makes re­com­ menda­tions to the justice minister about nominations for judicial appointments. Applicants for a judicial position send their application to the Judicial Appointments Council who nominates one, and only one, candidate for each appointment. The law presupposes that the justice minister follows the recommendations of the Council, but in exceptional cases in which the minister does not want to accept the recommendation of the Council, the Parliamentary Legal Affairs Committee must be notified. The Judicial Appointments Council has six members: a Supreme Court justice who is chairperson, a high court judge, a district court judge, an attorney, and two public representatives. The Judicial Appointments Council guarantees that the courts will have the final say on judicial appointments. The Constitutional Act does not contain an explicit provision about the courts’ power to adjudicate the constitutionality of acts, and their competence in such matters refer back to supreme court practices from around 1920. Today, this kind of constitutional testing is done frequently, with only a single case in 1999 resulting in the direct denial of a legal provision (U 1999.841 H). In that particular case (known as the Tvind Case), the Folketinget and the government had passed a bill that denied a number of schools the possibility of receiving government funding. The Supreme Court ruled that this was in violation of Section 3 of the Constitutional Act about the separation of power since the Parliament and government had in effect given a court ruling. The Supreme Court’s judicial review of specific legislation has traditionally been quite cautious, since in order for a law to be overturned by the Supreme Court, its variance with the Constitution has to be of significant certainty. This is undoubtedly still true of a number of constitutional provisions; however, it is just as important to note that specific legislation will only be deemed unconstitutional on rare occasions as the Constitutional Act only places minimal restrictions on the competence of the legislative power. It should also be mentioned that the Supreme Court’s review of the constitutionality of acts of Parliament is characterized by the fact that the Supreme Court, unlike some constitutional courts in other countries, has not engaged in dynamic and creative interpretations of the Constitution. The Danish Supreme Court has instead kept its feet on the ground and focused on common legal principles of interpretation based on the wording in the Constitutional Act and the meanings that can be presumed to be behind the formulations. This has prevented the politicization of judicial appointments in Denmark.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   25

Constitutional Rights As mentioned at the beginning of this chapter, the Danish Constitution is organized around two types of constitutional rights. These are the political rights of freedom of speech, freedom of assembly, and freedom of association; and the personal rights concerning provisions about personal freedom and the inviolability of home and property rights. In addition, one finds provisions about freedom of religion, the right to public assistance, and free education in public schools. These provisions are all based on the first Constitutional Act from 1849 and have only been expanded slightly in the various constitutional amendments. In this way, Denmark stands out among most European countries, including the Nordic countries, whose catalogues of constitutional rights are much more wide-ranging. The reason for this is partly that most of these countries have amended their constitutions over the past few decades, during which there has been a great deal of focus on and interest in many different kinds of constitutional rights. The Danish Constitution is very difficult to change, remaining unchanged since 1953 as a result, and so this ‘constitutional wave’ has not affected the Danish Constitutional Act. The constitutional provision about freedom of speech contains an absolute pro­hib­ ition on censorship, whereas the protection of the content of speech is relegated to the fact that responsibility for speech can be placed by means of a court ruling. In the same way, the provision about freedom of association establishes that associations cannot be annulled by the government but generally require a court ruling. The freedom of assembly is even more far reaching since citizens have a constitutional right to assemble unarmed. The Constitution protects personal freedom in particular by guaranteeing that an individual must come before a judge who will determine whether he or she will remain imprisoned within twenty-four hours of being arrested. The objective of this provision was and is to prevent government misuse of the criminal justice system. Likewise, in principle, house searches, seizure and examination of private documents, etc. have to be preapproved by a judge. However, legislation can and often does make exceptions, for instance, when it comes to attempts by the authorities to ensure compliance with tax laws, food laws, and the like. Property rights are specifically protected in such a way that in the case of expropriation of property, it is ultimately left up to the courts to decide whether or not the owner has received complete compensation. In an international context, Denmark has joined a number of human rights treaties since the end of the Second World War. Some of the most significant are the European Convention on Human Rights and the UN’s two human rights conventions about pol­it­ ical rights and economic and social rights. By signing treaties about human rights, the Danish state has committed itself to live up to these treaties. Generally, Danish citizens cannot rely directly on these treaties. However, if someone ends up in trial, they will get relatively close since the courts will refer to the fact that Danish laws have to be understood to the greatest extent possible in accordance with the treaties that Denmark has

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

26   Jens Peter Christensen joined. This is different for the European Convention on Human Rights since this treaty was introduced and approved as a bill in 1992, and thus, it functions as direct law. The law on the European Convention on Human Rights regulates a number of issues that are not regulated by the Constitution, as well as a number of the same issues such as freedom of speech, freedom of association, and freedom of assembly. However, as a law, it only has the status of a law and cannot be interpreted into the Constitution. Imagine, for instance, if the Supreme Court interpreted Section 77 in the Constitutional Act in accordance with Article 10 of the Human Rights Convention which also deals with freedom of speech but in a much broader scope. In doing so, the judges would suddenly change the content of the Danish Constitutional Act without adhering to the complicated procedures for amending the Constitution as outlined in Section 88 of the Constitutional Act. Moreover, the judges at the European Court of Human Rights in Strasbourg regularly interpret new content into the provisions of the Convention based on something called ‘dynamic’ interpretation but which, in a Danish context, has been referred to in less generous terms as ‘legal free-hand drawing’. In other words, if the Supreme Court interpreted the Constitution in accordance with the Human Rights Convention, it would mean that Denmark would delegate constitutional power to the judges of the Human Rights Courts in Strasbourg. Therefore, the Danish Supreme Court has never reinterpreted the Constitution in the light of convention provisions, but it has instead maintained a clear distinction between the Constitution and the treaties.

The Constitution that Never Changes The Danish Constitutional Act has been referred to as the ‘constitution that never changes’ (Christensen 2002: 99). This is not entirely accurate given that the Constitution has been amended a few times. However, it is true that such amendments have been rare, and there are no expectations for additional amendments in the near future. There are two reasons for this. The first reason is the complicated amendment procedure outlined in Section 88 of the Constitution. The practical implications of this section are that it requires very broad support in both the Folketinget and among voters. The second reason is that in terms of practical politics, constitutional politics are politics of necessity. Historically, the only time the Constitution has been amended has been when it blocked policies considered necessary by a substantial parliamentary majority. Currently, there are no pressing matters requiring a constitutional amendment. As mentioned, the constitutional framework outlined in the Danish Constitutional Act for political life in Denmark is simple and has allowed ample opportunity for practical parliamentary practices to unfold in accordance with the changing times and without any constitutional crises. In some cases, the constitutional provisions about personal and political rights and freedoms are a bit flimsy, but in practice, this does not prove to be a problem. The legal system is stock full of statutory rights, and national rules are

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/02/2020, SPi

The Constitution   27 supplemented across the board by EU provisions and the European Convention on Human Rights. There is no reason to expect a constitutional amendment in the foreseeable future. The major parties have no desire to open up the Pandora’s Box that an amendment process could very well turn out to be. Political life presents many challenges on a daily basis, but the Danish Constitution is not one of them.

References Christensen, Jens Peter (1993). ‘Debatten om grundlovsændring i forfatningsretlig belysning’, Ugeskrift for Retsvæsen, 127/1: 1–12. Christensen, Jens Peter (2002). ‘Norm og praksis under grundloven som aldrig ændres’, in Eivind Smith, ed., Grundlagens makt. Konstitutionen som politiskt redskap och som rättslig norm. Stockholm: SNS Förlag, 99–115. Christensen, Jens Peter (2017). Grundloven. Atten fortællinger, 3rd ed. Copenhagen: Gyldendal. Christensen, Jens Peter, Jørgen A. Jensen, and Michael H. Jensen (2016). Dansk Statsret, 2nd ed. Copenhagen: Jurist- og Økonomforbundets Forlag. Damgaard, Erik (1990) (ed.). Parlamentarisk Forandring i Norden. Oslo: Oslo University Press. Elklit, Jørgen (2020). ‘The electoral system. Fair and well-functioning’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 56–75. Folketinget (n.d.). ‘My Constitutional Act’, https://www.thedanishparliament.dk/∼/media/ pdf/publikationer/english/my_constitutional_act_with_explanations.ashx (accessed 21 January 2019). Ross, Alf (1959). Dansk Statsforfatningsret I. Copenhagen: Nyt Nordisk Forlag. U 1998.800 H. Ugeskrift for Retsvæsen, 1998, 800–71. U 1999.841 H. Ugeskrift for Retsvæsen, 1999, 841–51. U 2010.1547 H. Ugeskrift for Retsvæsen, 2010, 1547–75. White Paper (1571/2018). Betænkning afgivet af Udvalget om undersøgelseskommissioner. Undersøgelseskommissioner og parlamentariske undersøgelsesformer. Copenhagen: Ministry of Justice, http://www.justitsministeriet.dk/sites/default/files/media/Pressemeddelelser/ pdf/2018/251018_-_betaenkning_1571_2018.pdf Worre, Torben (1982). Det politiske system i Danmark, 4th ed. Copenhagen: Akademisk Forlag.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

chapter 3

Gr een l a n d, th e Fa roe Isl a n ds, a n d Denm a r k Unity or Community? Ulrik Pram Gad

From Margins to Centre Stage All nationalisms involve not only collective memories but also forgetting (Renan 1882/1992). The Danish nation state has largely forgotten its past and presence as a ‘conglomerate state’ (Bregnsbo 2004), and scholarship is complicit. Review articles covering the Nordic region often relegate autonomous territories and even sovereign Iceland to a footnote. In schoolbook maps, little boxes with the Faroe Islands and a downscaled Greenland appear, inserted in the margins without explanation. However, if mapping the world in polar projection (Figure 3.1)—which became urgent to military strategists with the advent of the nuclear age—it becomes obvious that the Kingdom of Denmark is not (only) a tiny plug in the Baltic bottleneck. Recently, the ‘scramble for the Arctic’ (Nuttall and Dodds 2016) set in motion by changes in global balances of climate, power, and economy has made its way to the Danish public, sparking renewed attention towards long-forgotten complications as to just what ‘Denmark’ is. This chapter identifies the most important characteristics and dynamics of the Danish ‘Community of the Realm’, Rigsfællesskabet, with Kalaallit Nunaat (Greenland) and Føroyar (the Faroe Islands). The main message is that every time Danish actions, reactions, or passivity can be taken to imply that Denmark tries to keep the Realm for itself, the existence of the Community is put at stake. The chapter is structured as three narratives of the Community in terms of constitutional status, sociological aspirations, and geopolitical context. The first section sets the stage by introducing the constitutional status of Greenland and the Faroes, the struggles that produced and contest it, and the linguistic games played to keep it together. The second section identifies three phases— analytically distinct but historically entangled—in the quest for equality among the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   29 Arctic Ocean

CANADA

RUSSIA North Pole Thule

Barents Sea

GREENLAND

Labrador Sea

Svalbard NORWAY

Norwegian Sea DEN SWE

ND

FAROE ISLANDS

A NL

ICELAND

Atlantic Ocean

Murmansk FI

Tasiilaq

Greenland Sea Ittoqqortoormiit

NO RW AY

Nuuk

Missile trajectory Kola Peninsula − New England

Figure 3.1  When seeing Greenland and the Faroe Islands between North America, Scandinavia, and Eurasia in polar projection, Denmark disappears (map adapted from geology.com).

three parties of the Community of the Realm and present examples of how they failed. The third section identifies a series of diplomatic games played: first, games played by Denmark to uphold sovereignty under difficult geopolitical conditions, and later, games played by Greenland, the Faroes, and Denmark more or less in concert to allow the two small polities independent international agency. The conclusion draws together a prognosis for how these games are likely to cross what has so far served as domestic bound­ar­ ies into uncharted territory, even if international conditions are increasingly difficult.

A Constitutional Oxymoron During the preparations for revised arrangements for the Faroe Islands and Greenland in the wake of the Second World War, the government asked the country’s preeminent legal mind Alf Ross for advice. He took the opportunity to describe Rigsfællesskabet—the commonly used label for the fact that the Kingdom of Denmark also includes islands, small and large, in the North Atlantic—as a legal nullity (cf. Harhoff 1993: 73) since ‘Community [fællesskab] exists among parties of equal status; unity [enhed] exists among parts and totality’ (quoted in Spiermann 2007: 11, own translation). Official legal text long avoided using the label, and the website of the Danish Prime Minister’s Office still insists on the ‘Unity of the Realm’ in its English version (Prime Minister’s Office n.d.). Nevertheless, Rigsfællesskabet remains the conceptual framework for popular im­agin­ations and practical arrangements as well as for legal struggles and political

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

30   Ulrik Pram Gad debate over the connections between Greenlanders, Faroes, and Danes. Literally, the Danish version translates into English as ‘Community of the Realm’ like the Faroese ríkisfelagsskapur, with felagsskapur connoting solidarity and companionship in contrast to the alternative translation samveldið (‘commonwealth’ or community of states), which was chosen to characterize the early incarnations of what is now the European Union. The Greenlandic version, naalagaaffeqatigiinneq, goes even further: naalagaq (the one who decides) is qatigiinneq, that is, something you do together with someone else (Lennert 2006: 1, n. 2). The ambiguity and polyvalency of the concept appears to facilitate convivance. The remaining part of this section will introduce the historical background for the current constitutional state of the (comm)unity and characterize the formal status of Greenland and the Faroe Islands. From the point of view of Danish constitutional history, Greenland and the Faroe Islands came under the Danish crown when it was fused with the Norwegian crown in the fourteenth century. The Norse had settled in both places and trade was conducted out of Western Norway. However, contact with the settlers in Greenland died out—and it later appeared that so had the settlers themselves. Greenland was recolonized from 1721, when Norwegian Hans Egede was sent by the King to save the souls of the Norse who had had no chance to hear about the Reformation. Since no Norse were to be found, Egede took upon himself to christen the Inuit who had migrated to the island. When Norway proper was lost to Sweden as part of the settlement after the Napoleonic wars, trade with and administration of Greenland and the Faroe Islands were transferred to Copenhagen. As discussed in detail below, the Faroe Islands were granted autonomy under the label ‘home rule’ in 1946, whereas Greenland was first decolonized by integration with the 1953 constitutional reform and achieved ‘home rule’ in 1979, a status that was transformed to ‘self-rule’ in 2009. Meanwhile, each community retains two members of the Danish Parliament (Folketinget) with equal status to other MPs and full voting rights in all matters (Pedersen 2020; Elklit 2020). Official interpretations of Danish constitutional law have insisted on treating the two autonomies separately but similarly as powers devolved from a sovereign centre (Christensen 2020). In contrast, the constitutional statuses of both Greenland and the Faroe Islands have been contested in slightly different ways referring to the distinct historical trajectories of submission to Copenhagen. Some Faroese historians and lawyers have argued that the Faroese constitute a distinct nation that never accepted transfer of sovereignty from the Norwegian to the Danish crown (Spiermann 2007: 47, fn.62). Greenlandic critique has c­ entred on whether the procedure culminating in the formal integration of Greenland in 1953 lived up to UN standards prescribing free and informed consent. Objections have particularly been directed to the fact that the alternatives to integration discussed in the UN decolonization processes—independence and free association—were not presented to the Greenlanders (Kleist 2019). Activist legal scholarship (Harhoff 1993; Spiermann 2007) argues that by now, the home and self-rule arrangements are not just irrevocable for pragmatic political reasons (Christensen 2020) but that they have a constitutional status which protects them against unilateral Danish intervention. In other words, traditional legal interpretation would categorize the Greenlandic and Faroese

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   31 governments as devolved, whereas activist legal scholarship and political and diplomatic practice would point towards a categorization of Denmark’s involvement with the islands as a federacy (Stepan 2013: 244; Justinussen 2019). Since the 2009 law on self-rule includes a pro­ced­ure for how Greenland may declare independence, contestations have concentrated on competencies in foreign and security politics. In continuation, activist legal scholarship (Spiermann 2007: 120–3) supports increased independent international agency for Greenland and the Faroe Islands by reference to their independent status as peoples under international law, against claims from the Danish government of a ­constitutionally protected prerogative over external affairs.

Competencies Reaching into Core State Functions The autonomies’ spheres of competence are wide compared to the arrangements between other overseas territories and their metropole states. Greenlandic and Faroese authorities have ‘taken over’ (as Danish law wants it; Greenlanders and Faroese prefer ‘taken home’) almost all domains of domestic affairs. Most remaining affairs are listed as eligible for takeover, including core state functions such as police, prosecution, courts, the penal system, passports, aliens, and border control. In the face of each Greenlandic and Faroese demand for increased autonomy, the official Danish position has been that it would violate the Danish Constitution—until one day, the government changed its position and taking home minerals extraction and police was suddenly possible as was involvement in foreign affairs and formal recognition of Greenlandic self-government under international law. Legal scholars debate whether official Danish interpretation of constitutional law has indeed constituted consistent limits (Danielsen 2011) or whether it is all down to politics (Larsen 2011). Only ‘Constitution; Citizenship; Supreme Court; Foreign, security and defence policy; Foreign exchange and monetary policy’ are ex­pli­ cit­ly exempted from takeover. Even within foreign policy, advanced procedures for independent agency and involvement have been codified. The governments of Greenland and the Faroe Islands may ‘on behalf of the Realm’ conclude agreements with foreign states and international organizations within the ‘fields of responsibility taken over’. However, Denmark takes over the steering wheel if Denmark (proper) is also involved in the matter, if both Greenland and the Faroe Islands are involved, or if Denmark is a member of the international organization in question. In such matters— and other foreign affairs conducted by the Danish government on behalf of the islands— intensive consultations are prescribed. Indeed, a precedence has evolved that would make it inconvenient for a Danish minister of foreign affairs to discuss Arctic affairs with a foreign colleague without a Greenlandic counterpart present (Gad 2017a: 18). The asymmetries of what is in effect a Danish federacy, as well as the contestations that produced the asymmetries and the games which keep the federacy going, are products of two underlying factors: the quest for equality shaped by and into nationalism, and particularly for Greenland, a unique combination of geographical position and territorial and population size.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

32   Ulrik Pram Gad

Seeking Equality in Asymmetrical Relations Building a ‘Community’ only became necessary when a ‘Realm’ ceased to be legitimate on its own terms, i.e. with the advent of notions of popular sovereignty. Even then, hierarchies of race, civilization, and size pre-empted the need for equality among the constituent parties of the Realm. Originally, Greenlanders were considered racially different, perhaps masters of their particular ecology but unfit for civilization, and hence, in need of Danish protection (Marquardt 2009). The Faroese language was once considered an uncivilized dialect unfit for instruction (Gaini 2011). Reminiscences of both ideas live on in prejudice (sometimes internalized) claiming that Greenlanders are not really fit to show up at work every morning at 8 o’ clock as they are programmed by a culture orienting life towards nature (Brochmann and Hamann 1990: 68–71), or in the way the Danish authorities may present themselves as guardians protecting the less-than-civilized Faroese customs like the Grindadráp harvest of pilot whales (Adler-Nissen 2014). However, the Faroese and Greenlanders have learned from their Danish overlords that the ideal way to achieve equality in a diverse world is to be a culturally homogenous nation in possession of its own state (Gad 2017a: 41–7). The last hierarchy standing concerns size; counting only some 50,000 inhabitants, some Faroese and Greenlanders join many Danes in deeming their nations too small to independently run a modern state (Breum 2019). Table 3.1 offers an overview of the asymmetries of the Community of the Realm.

No Turning Back after the Second World War The Second World War cut off Iceland, the Faroe Islands, and Greenland, governed by  local authorities while in UK and US custody, from metropole Denmark under

Table 3.1.  Basic Asymmetries of the Community of the Realm 2017 Polity

Territory (km2)

Population GDP (mDKK)

GDP/cap. Block grant Block grant/ Block grant/ (tDKK) (mDKK) budget GDP

Faroe Islands

1,396

50,498

18,900

377

642

14%

3,5%

Greenland 2,166,086 55,860 Ice free: 410,449

15,309

326

3,722

59%

25%

Denmark

2,096,000 361

−4,364

−4%

−0,2%

42,933

5,806,015

Note: The annual lump sum transfers from the Danish state to the governments of the Faroe Islands and Greenland are known as ‘block grants’. GDP are PPP figures based on preliminary estimates from Danmarks Nationalbank, Statistics Greenland, and Hágstova Føroyar.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   33 German occupation. When reconnected, there was no turning back to established hierarchies; only programmes for further equality made sense. Globally, the war ­ delegitimized racial hierarchies as the basis for formalized political subjugation, and integral to the formation of the United Nations was the imperative to decolonize all non-self-governing territories. However, all three North Atlantic polities ended up taking steps towards different kinds of equality. Since 1918, Iceland had been a formally sovereign state; the treaty establishing a personal union with Denmark left only foreign affairs and the coast guard to Copenhagen. Occupied, Denmark could not take part in the evaluation that the treaty stipulated for 1940, and in 1944, Iceland declared itself an independent republic, emerging as a formally, fully equal member of the international community of states. The Faroe Islands briefly reached for the same kind of equality when a narrow plurality in a 1946 low-turnout referendum voted for secession; the speaker of the Faroese consultative assembly, Lögting, declared the establishment of ‘The Faroese Realm’, and a visiting warship seemed to signal British recognition by flying the Faroese Merkið flag rather than the Danish Dannebrog as courtesy ensign while docked in Tórshavn (Harhoff 1993: 60–1). Danish authorities intervened by dispatching its own warship, dissolving the Lögting (upon the request of a majority of its members), and calling a general election which (again) returned a majority against secession. When the dust settled in 1948, the Danish Folketinget passed the Faroe Islands Home Rule Act, transferring extensive legislative and executive powers in domestic affairs to Tórshavn. The Act also declared the Faroe Islands ‘a self-governing community of people [folkesamfund] within the Danish Realm’, a somewhat archaic-sounding novelty connoting something less than a people under international law but also something more distinct than the populations of other Danish islands. The process cemented Faroese politics in a party system distributing parties along two dimensions: one traditional left/right and one union/independence. Four parties still dominate, each occupying a quadrant in the formative matrix (Adler-Nissen and Gad 2014). Hence, the status of the islands remains contested. While secessionists have instigated a series of preparatory measures when part of coalition governments, elect­ or­al and parliamentary majorities have so far stopped the process from culminating in (another) declaration of independence, particularly when confronted with Danish counter-demands. In 2001, the Danish prime minister responded to a Faroese plan for a referendum on independence with the threat of terminating Danish subsidies on only one year’s notice. Another Danish prime minister used the same threat to splinter a co­ali­tion of pro-independence and con-independence parties that had agreed to formulate a Faroese constitution without proper reference to the Danish Constitution (Gad 2017a: 114). Regarding Greenland, the Danish reaction to developments after the Second World War was to decolonize by integrating the island and its population under the 1953 Constitutional Act as an equal part of Denmark. Greenlanders did not participate in the constitutional referendum. Substantially, the integration was followed by intensified modernization drives, following demands from the Greenlandic elite, but substantially devised in and implemented by the Ministry for Greenland and its two Copenhagen-

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

34   Ulrik Pram Gad based executive arms, the Royal Greenlandic Trade (KGH) and the Greenland Technical Organization (GTO). Results were dramatic, complex, and contentious. On the one hand, the average life expectancy almost doubled (Iburg et al. 2001) due to better health services, social service, and housing quality, and the number of Greenlanders pursuing education beyond primary school rose. On the other hand, what was formally de­col­on­ iza­tion felt like neocolonialism; the physical infrastructure framing the new welfare was built by Danish entrepreneurs and builders, more Danes were recruited to man the welfare institutions, and both infrastructure and institutions were largely built on Danish blueprints (Dahl 1986). The road to equality as Danish citizens was paved with inequality. Particularly offending was the ‘birth place criterion’ (Janussen 2017) introduced to protect the (less productive) Greenlandic labour market from adjusting to (higher) Danish salaries; people born in Greenland were paid less than colleagues with identical education and function recruited from Denmark. These experiences of discrimination and government from far away were instrumental in spurring demands for different kinds of equality.

Finding Inspiration for New Equalities The Faroes had been looking to Iceland, and now the obvious place to look for in­spir­ ation for Greenlandic equality was the Faroe Islands. The 1979 Greenland Home Rule Act upgraded the consultative Landsrådet to a legislative Landstinget and established the executive Landsstyre. To substantiate the formal transfer of powers, the monolithic KGH and GTO organizations were divided functionally (Skydsbjerg 1999), their personnel were transferred to Nuuk (or substituted with new recruits), and ambitious reforms were instigated, particularly of the all-decisive fishing industry (decentralizing production and decision making to cities and settlements along the coast) and the educational sector (prioritizing the Greenlandic language) (Gad 2017c). Already by 1992, much earlier than anyone had imagined (Skydsbjerg 1999), the home rule government had built a brand new bureaucracy from scratch in Nuuk and taken home not only political responsibility but also the practical organization of major services from hospitals and helicopters to grocery stores in next to 100 isolated settlements and a world-leading shrimp export business. However, the feeling of being equal again proved evasive. The new positions in Nuuk under Greenlandic political control were still overwhelmingly filled with Danes sporting degrees rather than Greenlanders lacking degrees but familiar with ‘Greenlandic conditions’ (Binderkrantz 2008). Infrastructure, business connections, and consumer demands kept pointing traffic and commerce towards Denmark (Gad 2017a: 117) and reinforcing existing ethnic hierarchies (Gad 2017b: 220). Danish paradigms, blueprints, and education kept informing regulation, organization, and solutions to Greenlandic problems (Jakobsen 2008)—often in surprisingly automatic ways. One example is the way in which the 2007 Danish fusion of municipalities (Houlberg and Ejersbo 2020)

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   35 was imitated in 2009 where eighteen Greenlandic municipalities were merged to just four. One was named Sermersooq, literally ‘over the ice’, since it fuses the capital Nuuk with the East Coast (where people speak a different language). Another named Qaasuitsup, spanning 1,600 kilometres of coast, was proudly declared to be the largest municipality in the world (Qaasuitsup Kommunia 2014: 8), only to be split in two as early as 2018 following popular demand and a referendum. Apart from continued Danish dominance in import of expertise and manpower, lack of revenue under home rule was a severe barrier felt to be hindering Greenland from achieving equality as self-governing: Greenland relies on a Danish lump sum subsidy. Crucially, Greenland’s rich subsoil resources, including both well-documented onshore mineral deposits and potentially immense offshore oil, remained state property under the 1979 Home Rule Act. Finally, the lack of recognition as a people eligible for selfgovernment under international law kept rankling. In 2009, a reformed version of home rule dubbed ‘self-rule’ reprioritized these barriers to Greenlandic equality. First, the list of domains which Nuuk may take home was extended (as detailed above) to include even core state functions which official jurisprudence had ruled out of question in 1979. However, financially, the tables were turned; since 1979, every takeover had been accompanied by an expansion of the Danish subsidies to Greenland, but after 2009, Greenland has to finance new expenses on its own. Second, to make it possible to even imagine substantial self-funded takeovers, Greenland was allowed to take home regulation, administration, and possible revenues from extractive industries. The Greenlandic government confidently touted oils and minerals as the means to build a self-supporting economy (Gad et al. 2018), and hence, achieve financial equality.

Eligible for Equal Sovereignty? Finally, the 2009 act staged two forms of equality between Denmark and Greenland. Symbolically, the act simulated a present equality. Even though the law was passed by the Danish Parliament, it included a preamble—highly unusual in a Danish tradition— explicitly ‘[r]ecognising that the people of Greenland is a people pursuant to inter­ nation­al law with the right of self-determination’ and explaining that ‘[a]ccordingly, the Act is based on an agreement between Naalakkersuisut [using the Greenlandic name of the Government of Greenland] and the Danish Government as equal partners’. However, in between these two sentences, the preamble postponed equality, intimating the motivation for the act to be ‘a wish to foster equality and mutual respect in the partnership between Denmark and Greenland’ (italics inserted). The end of the act follows up on the initial reference to international law by explicating that ‘Decision[s] regarding Greenland’s independence shall be taken by the people of Greenland’ and drawing up a roadmap for secession. Formally, then, the Danish and Greenlandic peoples are presently equal as eligible for sovereignty; however, equality as sovereign is postponed for the future. Even with this temporal staggering of equality, the preamble seems to place

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

36   Ulrik Pram Gad Greenland one step ahead of the Faroe Islands in terms of equality as the Faroes never got Denmark to issue a similar recognition of a foundation of future Faroese independence in international law (Prime Minister 2018). Legal formalities and material substance aside, most Danish politicians have learned from the debates leading up to and following 2009 to explicitly deny hierarchy whenever mentioning the Community of the Realm. However, they tend to have more difficulties explaining just what equality among unequal partners may specifically mean (Gad 2017a: 42–44, 116). Against this background, most Greenlandic political parties and parliamentarians remain committed to equality as sovereignty; what organizes the party system is rather how fast one wants to progress with formal sovereignty and what substantial developments need to be tended to before sovereignty can be sustainable (Gad 2014).

Geopolitical and Paradiplomatic Sovereignty Games In these negotiations of how to get to sovereign equality, diversifying dependence beyond Denmark has been central. Playing games with Denmark’s sovereignty has been key to upholding it at all (Gad 2017a: 67–72). I employ the ‘games’ metaphor here neither as the stylized games of rational choice game theory, nor to convey the impression that actions were playful much less that the consequences were entertaining. Rather, as developed in Adler-Nissen and Gad (2014), ‘sovereignty games’ are linguistic and practical games constitutive of human interaction (pace Wittgenstein) and made possible and necessary by the idea that a state needs sovereignty to exist. Geopolitical position, territorial size, and imperial history, however, have made for very different games for Greenland and the Faroe Islands.

Imperial and Cold War Sovereignty Games Greenland is key to controlling access to the North American continent for missiles from Eurasia in general and from the Kola Peninsula in particular. The United States took over Greenland in 1941 to keep the Germans out and facilitate US participation in the war in Europe, sending provisions for the population while leaving civilian government to the Danish authorities in place. After re-establishing connections in 1945, Copenhagen expected the Americans to leave, but the development of weapons technology made the northernmost part of Greenland essential to US nuclear strategy, both defensive and offensive. The result was the 1951 Defense of Greenland Agreement (Lidegaard 1999: 179–88, 334–5; Wivel 2020) which ‘[w]ithout prejudice to the ­sovereignty

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   37 of the Kingdom of Denmark’ over Greenland, allowed the US military to do whatever it wanted. The fact that Danish acceptance of US priorities extended to the storage and transportation of nuclear weapons, which the Danish government explicitly forbids in Denmark ‘proper’, became apparent in 1968 when a B-52 crashed into the ice-covered fiord by Thule Air Base, and finally clear in 1995 when a 1957 letter from the Danish Prime Minister to the US embassy in Copenhagen discussing the matter was declassified (DUPI 1997). In comparison, the Faroe Islands had less negative effects from both the Second World War and the Cold War. UK troops arrived in Tórshavn as the Germans occupied Denmark, and in 1945, they left behind the island’s first airport and a community that had experienced five years of locally based government as well as commercial relations beyond the realm. During the Cold War, the position occupied by the islands in the ‘GIUK’ gap (the straits between Greenland–Iceland–UK) made for technical installations vital to NATO’s defence against Russian nuclear bombers, submarines, and missiles. Personnel were limited and mostly Danish, but the military character of some of the installations and the exact role of the Americans dispatched were kept secret from the Faroese public and Parliament, causing some local uproar when facts were un­covered (Jensen 2014; Nielsen 2017). In contrast to the Faroese experience, the Greenland Defense Agreement stipulated that ‘every effort will be made to avoid any contact between United States personnel and the local population’. This provision was in line with Danish narratives legitimizing colonization, namely that the Inuit natives needed protection from outside interference and that Denmark provided a uniquely gentle and benevolent stewardship of the modernization process. Nevertheless, Denmark’s efforts to uphold sovereignty over Greenland vis-à-vis competing powers have, in certain instances, involved harsh interference in the lives of groups of Greenlanders (Gad 2017a: 69–70, 2017b). In 1925, 70 Greenlanders were transferred some 800 km north from Tasiilaq to what would be known as Ittoqqortoormiit. Like in Arctic Canada (Salter  2019), transfers were promoted to the Greenlanders as a way to reach better hunting grounds, but the decisions were actually made to secure sovereignty: North East Greenland was claimed by Norway. Ultimately, the exercise proved helpful when the International Court of Justice in The Hague ruled in favour of Denmark in 1935. Decades later, the UN decolonization process and US military strategy conspired to push Denmark to a similar population transfer. In early 1953, the US found it necessary to protect Thule Air Base by building a tactical nuclear missile silo right in the middle of a neighbouring Inuit settlement (and the Danish authorities found the ‘contact between United States personnel and the local population’ unfolding in less than dignified ways). Since the revision of the Danish Constitution was due to expand civil liberties to Greenlanders from June 1953—as a reply to UN demands for decolonization of non-selfgoverning territories—the Danish authorities found themselves in a hurry. Hence, in the late spring, the Greenlanders were told to vacate the premises within days and pack their belongings on their dog sleds. Replacement housing at the abandoned summer

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

38   Ulrik Pram Gad campground Qaanaaq 100 km to the north was only established during the following winter. No sooner than in 1999, a high court decision established that the removal of the hunters and their families had been forced rather than voluntary as claimed by the authorities. The number of US facilities across Greenland has gradually shrunk to one (Thule), and a series of procedures have been introduced to enhance Danish and Greenlandic involvement and insight into US military activities (Ministry of Foreign Affairs 2008). The steady stream of scandals emerging from Thule has continuously embarrassed Denmark’s image as a particularly benevolent protector of its Greenlandic subjects. Hence, the 2003 Itilleq agreement guaranteed Greenland insight and involvement in the foreign and security policy conducted by the Danish government on behalf of the island and its population. Most spectacularly, the Government of Greenland got to co-sign the 2004 Igaliku agreement amending the defence agreement alongside the United States and Denmark in exchange for acceptance of US plans to revamp the Thule radars for the national missile defence system (Kristensen 2005). Denmark symbolically upholds sovereignty by flying Dannebrog over Thule, by dispatching the Sirius dogs-led patrol across uninhabited North East Greenland, and by patrolling Greenlandic waters with a handful of vessels conducting fisheries control. But all planning for war places the defence of Greenland in US hands. In parallel, Greenland now flies its flag Erfalasorput at Thule, but its government has not prioritized its sparse resources to dispatch an official representative at the base, and ideas for Greenlandic involvement in Danish military ac­tiv­ities in Greenland remain jottings so detached from reality that only recently, someone thought of adding Erfalasorput to Dannebrog on the stanchions of the Sirius sleds (High Commissioner 2020: 10). In that sense, the games played around military affairs allow the representatives of a tiny people a certain insight while legitimizing Danish formal sovereignty and US military sovereignty over a huge territory.

European Sovereignty Games In other parts of the foreign policy of the Community of the Realm, the benefits of playing games with sovereignty fall more decidedly to the islands. Relations with the European Union provide instructive examples (Adler-Nissen and Gad 2014), particularly since the games played in relation to the Faroe Islands and Greenland take different forms. When Denmark voted to join the European Communities in 1972, the Faroese votes under home rule were counted separately. Since a majority voted against, the Faroes were exempted from Danish membership. In contrast, Greenland’s majority against accession was inconsequential; integrated as an ‘equal’ part, Greenland followed Denmark into the EC. The prospects of sovereignty over fisheries, vital to Greenland’s economy and development plans, moving further South from Copenhagen to Brussels were decisive for the demands for a home rule status equal to the Faroes’ (Rebhan 2016; Beach 2020). Aiming to replicate the Faroese example, home rule was introduced in 1979, and a second referendum in 1982 showed a majority for leave. However, when

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   39 Greenland did leave the EC in 1985, it ended up in a radically different position than the Faroes. The two positions involve very different benefits and drawbacks, inviting different games to be played with sovereignty. The Faroes remain among the polities least involved in European integration. As a tiny economy outside any larger formal frameworks for cooperation and trade, it has been difficult for the islands to move themselves up the agenda of Eurocrats; only in 1991 was a trade agreement with the EU concluded (Adler-Nissen 2014). Their position outside the EU framework makes for interesting games. One example, perhaps the most intriguing to standard conceptions of international agency, played out in 2013, when herring and mackerel started moving in new patterns in the North Atlantic. Denmark has left sovereignty over fisheries to Brussels, and Tórshavn has taken home its own fisheries. The two sides could not agree on quotas, and the EU banned Faroese fishing products from the EU and Faroese vessels from EU ports (including Denmark). In return, the Faroese Ministry of Foreign Affairs advertised that it would (in Denmark’s name) initiate proceedings against the EU (including Denmark) in the WTO (Lögmansskrivstovan 2013). The dispute with the EU was settled before WTO pro­ced­ ures began, but in the meantime, the Faroes had redirected most of its fisheries export to Russia. More benefits from their outsider status came in 2014 when the EU introduced sanctions on Russia following the annexation of Crimea. The Faroe Islands quickly became the no. 1 exporter of fish to Russia and now works towards a free trade agreement with the Russian-led Eurasian Economic Union (Breum  2018) to protect what amounts to 29 per cent of the islands’ exports. Contrary to the Faroe Islands, Greenland left the EC only to be associated with the communities through the well-established category of Overseas Countries and Territories (OCT). The OCTs remain outside the EU acquis, but they enjoy free movement of people and tax-free access for exports to the Single Market. The 1985 Greenland Treaty amended the Treaty of Rome to move Greenland from Danish membership to OCT status, but tax-free status only applies on the provision that ‘satisfactory’ agreements on fisheries cooperation are in place. The fisheries agreements involve the EU buying fishing quotas, for decades securing annual transfers from Brussels which matched what Greenland received from regional development funds while a member (Gad 2017a: 103–4). When the European Parliament was granted oversight powers, the price paid for fishing quotas from Greenland came under pressure; counting ‘paper fish’ that mainly existed as potential future stocks which might conceivably one day appear in Greenlandic waters, parliamentarians found that quotas could be bought cheaper in African waters. Nevertheless, Greenland and Denmark managed to keep the amount transferred from Brussels roughly constant by convincing EU members to sign a unique Partnership Agreement aiming to support the ‘sustainable development’ of Greenland (Gad 2017a: 82, 104). Contrary to most OCTs, Greenland’s relatively high GDP precludes eligibility for EDF funds. Particularly when contrasted to the Faroe Islands’ outsider position, it is clear that the OCT status, fisheries, and partnership agreements allow Greenland uniquely independent visibility and agency in relation to the EU for a nonsovereign, non-member polity (Gad 2017a: 85–100).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

40   Ulrik Pram Gad

Arctic Sovereignty Games Increased attention in the wake of changes to global climate as well as economic and security balances has awarded Greenland and the Faroe Islands more possibilities to connect beyond Denmark. Both governments deploy their own diplomatic staff (with Danish diplomatic passports) to Brussels and Reykjavik, Tórshavn also sends rep­re­sen­ ta­tives to London and Moscow, and Nuuk recently added Washington and have plans for Beijing. However, it has become increasingly clear that not all third parties are as apt as the EU at engaging in similar sovereignty games. Moreover, recent US attitudes towards the ‘scramble for the Arctic’ might tell Greenland that Eastern relief of economic dependence on Western countries is unacceptable. Greenland’s relations to China in particular have been complicated (Gad et al. 2018). When gearing up efforts to attract foreign investments in the extractive industries meant to finance independence, Greenlandic authorities realized that the global markets for raw materials had just plummeted with the 2007 financial crash. Soon, they began looking to China for investors driven by sufficiently long-term interests. Initially, the Chinese counterparts seemed to have some trouble deciding how to engage with this ‘Tibet-size’ autonomous entity; later, Greenlandic efforts—and the way they were facilitated by Danish diplomacy—appears to have persuaded Chinese organizations that it would be alright to do business with Greenland. The extension of the Chinese Belt and Road Initiative with a Polar Silk Road has possibly made a difference as well (Sørensen 2018). Recently, Chinese investors have taken steps towards realizing a couple of mining projects (Andersson et al. 2018). More controversially, Greenlandic authorities included the China Communications Construction Company (CCCC) among entrepreneurs prequalified to bid for the expansion of two airports in Greenland. However, the Danish Prime Minister unexpectedly stepped in to co-fund the economically dubious project as a way to keep Greenland out of what some in Copenhagen portrayed as a Chinese debt trap. A couple of years earlier, Denmark advertised the old US/Danish naval station at Kangilinnguit for sale only to withdraw the property from the market when a Chinese mining company showed interest in buying. Most Greenlandic politicians are content with securing Danish funds for the airports. However, particularly when seen in conjunction, the two interventions nurture suspicions that even while trying to keep up a facade of selflessly facilitating Greenland’s diversification of its dependence, Denmark works to keep the dependence for itself to uphold sovereignty over Greenland. Such Danish efforts might be made easier by a re-securitization of the wider Arctic. Russia seems to prioritize economic development of its Arctic, and to that aim, it needs Western technology and Chinese investments. Meanwhile, Russian efforts to enhance centrifugal dynamics in Europe and the West are well documented. The Faroese fisheries exports—and the Danish acquiescence—play well into both Russian strategies. Danish authorities have had trouble convincing Nuuk that the worries about Chinese involvement in Greenland originated in Washington. However, with US Secretary of State Mike Pompeo’s statements at the Arctic Council Ministerial Meeting in May 2019, the Trump administration’s approach to alliance politics arrived in the Arctic. While

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   41 placing Russian demands to control the Northern Sea Route along its Siberian coast in ‘a pattern of aggressive Russian behaviour here in the Arctic’, his main worry was its inclusion in China’s Polar Silk Road. Thus, he conjured up to images: First, an ‘Arctic Ocean . . . transform[ed] into a new South China Sea, fraught with militarization and competing territorial claims’. Second, an image in which ‘all nations, including non-Arctic nations, . . . have a right to engage peacefully in this region[. . . in] free and fair competition, open, by the rule of law’, but since ‘all the parties in the marketplace have to play by those same rules [and t]hose who violate those rules should lose their rights to participate in that marketplace, [r]espect and transparency are the price of admission’ (Pompeo 2019). It was understood in whose hands admission control would best be put, as ‘American leadership stands in stark contrast with the Chinese and Russian models’ (Pompeo 2019).

Conclusion The ‘Community of the Realm’ remains a constitutional oxymoron connoting both imperial hierarchy and communal bonds. Its three constituent polities share an ideal of a world of nations, each homogenous in ethnic terms, and each in command of their own state. Step by step, Denmark has met some Faroese and Greenlandic demands, but the equality aspired for proves itself elusive. Over time, the games played to protect Danish sovereignty have changed from somewhat heavy-handed, if never bloody, Danish imperialism to increasingly sophisticated and collaborative linguistic and diplomatic games. Given the nation state ideal exported from Copenhagen to its dependencies, the narratives most legitimate in the North Atlantic describe the Community as an ever-looser union. Hence, Denmark greatly enhances the viability of the Community by explicitly embracing its dissolution—in the form of Greenlandic and Faroese independence—as its ultimate goal. In contrast, whenever Danish actions, reactions, or passivity can be taken to imply that Denmark tries to keep the Realm for itself, the existence of the Community is put at stake. Looking ahead, the domestic side of these games is likely to be easier than the international side. After initially dismissing the whole enterprise, the Danish government has changed its position, accepting that separate Faroese and Greenlandic constitutions are ac­cept­ able as long as they do not contradict the Danish Constitution, Grundloven. Nevertheless, after some initial confusion, the Greenlandic constitutional committee is headed towards drafting a constitution for a sovereign Greenlandic state (Schultz-Nielsen 2019). The adoption of such a text and a parallel Faroese one will formally kill off the Community of the Realm as we constitutionally know it. However, that will hardly be the end of Danish-Greenlandic-Faroese relations. Quite the contrary, a ‘Community’ immediately resurrected after secession, re-constituted as less of a ‘Realm’, is probably the most durable version. Different historical narratives producing different legal bases for demands have mostly kept Greenlandic and Faroese efforts separate. The Faroese have long looked to the 1918 Icelandic personal union (Hauge  2018), while the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

42   Ulrik Pram Gad Greenlanders, inspired by other decolonizing archipelagos, have fixed on ‘free association’ as a better fit (Kleist 2019). Increasingly, however, the specific queries and suggestions have appeared in tandem. The next step could be that Greenland and/or the Faroe Islands leave the Danish state, only to associate from outside. In the negotiations leading to this rearrangement, constitutional law, day-to-day foreign policy, and finances will be fought over but ultimately resolved in line with Danish pragmatism. Deal breakers will rather be either the organization of security politics or a hard choice to be made by the islanders. The security related obstacle pertains more severely to Greenland than to the Faroes. After Secretary Pompeo at Rovaniemi had drawn up his image of a region in conflict and promised to send more naval ships, he concluded his Arctic offensive by sending a parttime diplomat to Nuuk. The traveling First Secretary presented his task as ‘forging bonds between people and expanding our economic and trade bonds’ (Wille and Lindstrøm 2019). However, President Trump’s surprise attempt to cut through sovereignty games by simply buying Greenland might have served to explain to the Government of Greenland the bonds put on its sovereign choice of business partners by US security doctrines. This could very well start a new conversation in Greenland about its position in geopolitics. So far, geopolitics in Greenland has been discussed either as threats coming from outside in the form of US military facilities drawing attention from Russian nukes, or as opportunities to diversify dependence by engaging rising powers. Chances are, the conversation will now have to balance the risks and opportunities involved in not picking sides. If the resident superpower explicitly insists that it will not accept that Greenland keeps all options open, Denmark might even reappear as a convenient buffer between Greenland and the real(ist) world. The hard choice comes from the precedence in international jurisprudence making separate citizenship the fine line deciding eligibility for independent membership of the UN and some, but far from all, other international organizations, extending from the intergovernmental organizations managing North Atlantic fish stocks to the private International Olympic Committee. Freely associated states whose citizens also enjoy citizenship of a metropole state are not allowed membership; those who do not can attend both the UN general assembly and the Olympics under their own flag (Kleist 2019). Only time will tell whether the Greenlanders and the Faroese value these flags more than access to social services in Denmark and residency and work in the European Union allowed by a Danish passport.

References Adler-Nissen, Rebecca (2014). ‘The Faroe Islands’, Cooperation & Conflict, 49/1: 55–79. Adler-Nissen, Rebecca, and Ulrik  P.  Gad (2014). ‘Post-imperial sovereignty games in the Nordic region’, Cooperation & Conflict, 49/1: 3–32. Andersson, Patrik, Jesper Zeuthen, and Per Kalvig (2018). ‘Chinese mining in Greenland: Arctic access or access to minerals?’, Arctic Yearbook 2018: Special Section: China & the Arctic, 102–17.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   43 Beach, Derek (2020). ‘Referendums in Denmark: Influence on politics’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 400–16. Binderkrantz, Anne (2008). ‘På danske hænder?’, Politica, 40/2: 155–79. Bregnsbo, Michael (2004). ‘Fra konglomeratstat til enhedsstat’, paper presented to a seminar on The Multicultural Denmark of the long 18th century, Hillerød, 25–26 April. Breum, Martin (2018). ‘Russian fish money keeping Faroes out of EU sanctions’, EUobserver, https://euobserver.com/foreign/142847 (accessed 26 June 2019). Breum, Martin (2019). ‘Bor der virkelig mennesker nok i Grønland til at drive en stat?’, djøfbladet, https://www.djoefbladet.dk/artikler/2019/1/hvor-mange-mennesker-skal-der-til-atdrive-en-stat.aspx (accessed 26 June 2019). Brochmann, Helene, and Bente Hamann (1990). Forvaltning i Forandring: om kultur, rationalitet og bureaukrati i den grønlandske centraladministration. Nuuk: Atuakkiorfik. Christensen, Jens Peter (2020). ‘The constitution’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 9–27. Dahl, Jens (1986). Arktisk Selvstyre. Copenhagen: Akademisk. Danielsen, Jens (2011). ‘Grønlands selvstyre og Danmarks riges Grundlov’, Juristen, 93/1: 9–18. DUPI (1997). Grønland under den kolde krig: dansk og amerikansk sikkerhedspolitisk 1945–68. Copenhagen: DUPI. Elklit, Jørgen (2020). ‘The electoral system. Fair and well-functioning’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 56–75. Gad, Ulrik  P. (2014). ‘Ahead of snap elections, Greenland’s independence ambitions could open a window for closer co-operation with the EU’, EuroPP blog, http://bit.ly/1wIzsTL (accessed 26 June 2019). Gad, Ulrik P. (2017a). National Identity Politics and Postcolonial Sovereignty Games: Greenland, Denmark, and the European Union. Copenhagen: Museum Tusculanum. Gad, Ulrik P. (2017b). ‘Sex, Løgn og Landingsbaner: Thule set relationelt’, Grønland, 65/3: 216–41. Gad, Ulrik P. (2017c). ‘What kind of nation state will Greenland be? Securitization theory as a strategy for analysing identity politics’, Politik, 20/3: 104–20. Gad, Ulrik P., Naja Graugaard, Anders Holgersen, Marc Jacobsen, Nina Lave, and Nikoline Schriver (2018). ‘Imagining China on Greenland’s road to independence’, Arctic Yearbook 2018: Special Section: China & the Arctic, 6–28. Gaini, Firouz (2011). ‘Cultural rhapsody in shift’, in Firouz Gaini, ed., Among the Islanders of the North. Tórshavn: Fróðskapur, 132–63. Harhoff, Frederik (1993). Rigsfællesskabet. Aarhus: Klim. Hauge, Mads (2018). ‘Kongehuset og Færøerne’, unpublished manuscript, https://xn— slgtogkrone-b9a.dk/wp-content/uploads/2018/05/Kongehuset-og-F%C3%A6r%C3%B8erne. pdf (accessed 26 June 2019). High Commissioner (2020). ‘Indberetning fra rigsombudsmanden i Grønland’, 6 February, https://www.ft.dk/samling/20191/almdel/GRU/bilag/32/2147801.pdf (accessed 3 March 2020). Houlberg, Kurt, and Niels Ejersbo (2020). ‘Municipalities and regions. Approaching the limit of decentralization?’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 141–59. Iburg, Kim  M., Henrik  B.  Hansen, and Peter Bjerregaard (2001). ‘Health expectancy in Greenland’, Scandinavian Journal of Public Health, 29/1: 5–12.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

44   Ulrik Pram Gad Jakobsen, Mads L.F. (2008). ‘Et blindt fokus på Danmark?’, Politica, 40/2: 134–54. Janussen, Jakob (2017). Analyse af fødestedskriteriet. Nuuk: Forsoningskommissionen. Jensen, Bent (2014). Militære institutioner og anlæg på Færøerne under Den Kolde Krig. Copenhagen: Gyldendal. Justinussen, Jens C.S. (2019). ‘Rigsfællesskabet i et føderalt perspektiv’, Politica, 51/4: 441–68. Kleist, Mininnguaq (2019). ‘Grønlands udenrigspolitik og internationale relationer’, Politik, 22/1: 84–101. Kristensen, Kristian S. (2005). ‘Negotiating base rights for missile defence’, in Bertil Heurlin, and Steen Rynning, eds, Missile Defense. London: Routledge. 183–208. Larsen, Bárður (2011). ‘En håbløs søgen efter forfatningsrettens sikre fundament’, Juristen, 93/4: 128–35. Lennert, Karen H. (2006). ‘The Danish Interests in the community with Greenland’, unpublished Master’s thesis, Aalborg University. Lidegaard, Bo (1999). I Kongens Tjeneste. Copenhagen: Samleren. Lögmansskrivstovan (2013). ‘Faroe Islands takes the European Union to the WTO’, http:// www.mfa.fo/Default.aspx?ID=13626&Action=1&NewsId=5395&PID=23631 (accessed 13 March 2014). Marquardt, Ole (2009). ‘H.J.  Rink’, in Ole Høiris, ed., Grønland—en refleksiv udfordring. Aarhus: Aarhus University Press, 129–54. Ministry of Foreign Affairs (2008). ‘UGF alm. del – Svar på Spørgsmål 7’, https://www.ft.dk/ samling/20081/almdel/ugf/spm/7/svar/591827/629653.pdf (accessed 26 June 2019). Nielsen, Jens  P. (2017). ‘Færøerne’, in Danmark under den kolde krig, http://koldkrig-online. dk/anlaeg/faeroerne (accessed 26 June 2013). Nuttall, Mark, and Klaus Dodds (2016). The Scramble for the Poles. Cambridge: Polity. Pedersen, Helene H. (2020).‘The Parliament (Folketinget)’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 88–106. Pompeo, Mike (2019). ‘Looking north: Sharpening America’s Arctic focus’, remarks to Arctic Council Ministerial Meeting, https://translations.state.gov/2019/05/06/lookingnorth-sharpening-americas-arctic-focus (accessed 26 June 2019). Prime Minister (2018). ‘Færøudvalget spm 8’, https://www.tjodveldi.dk/spurningar/2018/4/18/ frudvalget-spm-8-om-pramblen-til-den-grnlandske-selvstyrelov-forpligter-danmarktil-folkeretligt-at-respektere-grnlands-eksterne-selvbestemmelsesret-til-at-etableregrnland-som-selvstndig-stat (accessed 5 June 2019). Prime Minister’s Office (n.d.). ‘The unity of the realm’, http://stm.dk/_a_2752.html (accessed 26 June 2019). Qaasuitsup Kommunia (2014). Kommuneplani 2014–26. Siunnersuut. Ilulissat: Qaasuitsup Kommunia. Rebhan, Christian (2016). North Atlantic Euroscepticism. Tórshavn: Froᵭskapur. Renan, Ernest (1882/1992). Qu’est-ce qu’une nation? Paris: Presses-Pocket. Salter, Mark (2019). ‘Arctic security, territory, population’, International Political Sociology, 13/4: 358–74. Schultz-Nielsen, Jørgen (2019). ‘Forfatningsarbejde: Omstridt punkt er skrottet’, Sermitsiaq, https://sermitsiaq.ag/node/212571 (accessed 26 June 2019). Skydsbjerg, Henrik (1999). Grønland. 20 år med hjemmestyre. Nuuk: Atuagkat. Spiermann, Ole (2007). Det danske rige i forfatningsretlig belysning. Copenhagen: Jurist- og Økonomforbundets Forlag.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Greenland, the Faroe Islands, and Denmark   45 Stepan, Alfred (2013). ‘A revised theory of federacy’, in Joanne McEvoy, and Brendan O’Leary, eds, Power Sharing in Deeply Divided Places. Philadelphia, PA: University of Pennsylvania Press, 231–52. Sørensen, Camilla T.N. (2018). ‘China is in the Arctic to stay as a Great Power’, Arctic Yearbook 2018: Special Section: China & the Arctic, 43–58. Wille, Andreas, and Merete Lindstrøm (2019). ‘Diplomaten Sung Choi skal knytte Grønland og USA tættere sammen’, KNR, https://knr.gl/da/nyheder/diplomaten-sungchoi-skal-knytte-gr%C3%B8nland-og-usa-t%C3%A6ttere-sammen (accessed 26 June 2019). Wivel, Anders (2020). ‘In war and peace. Security and defence policy in a small state’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 453–69.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

chapter 4

The Mona rch Head of State and National Symbol Jes Fabricius Møller

The monarch is the head of state. According to the most recent textbook on constitutional law, the role of the monarch is ‘of a purely symbolic and ceremonial character’, and the monarch has no independent competencies in matters of state (Christensen et al. 2015: 59). During the twentieth century, the functions of the monarch stabilized into a predictable relation and division of labour between Parliament, government, and head of state, equal to the heads of state in the other Nordic countries. According to the Constitution of 5 June 1849, which abolished absolute monarchy and introduced restricted monarchy, the king enjoyed a wide range of prerogatives. The role of the monarch has since changed considerably more than the wording of the Constitution, which has remained unchanged for the most part. According to Section 3, the king has executive power and shares legislative power with Parliament. According to Section 12—added in 1855—the king has ‘the highest authority over all matters of the realm and executes it through his ministers’, and his powers are only limited by what is stated in the Constitution itself. Taking into consideration that Denmark today is a parliamentary democracy, the functions of the head of state are to be executed on the basis of a careful reinterpretation of the wording and the intentions of the Constitution. Although the monarch has no political power, he or she is still considered the highest-ranking person in the country, a status that is confirmed symbolically (i.e. on coins and stamps) and at official events and ceremonies. This raises the question of how the monarch fits into the system of the sep­ ar­ation of power, according to which the organs of the state are equal and independent of each other. According to this principle, ‘it is incorrect to single out one of the organs of the state as superior to the others’ (Sørensen 1973: 58). The answer to this question is that all political and judicial power is divided between government, Parliament, and the courts; the monarch has neither political nor judicial power, and as such, is beyond the separation of power. The function of the monarch is to be a ‘symbol of the unity of the state’ (Andersen 1954: 112). The monarch does not define or incarnate but represents the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

The Monarch: Head of State and National Symbol   47 state. Therefore, when the president of the Supreme Court, the speaker of Parliament, or the prime minister pays their respects to HM Queen Margrethe II, it is a sign not of subordination but of respect for the state they all serve. The monarch is elevated above political struggle and thus liberated from the execution of power. The monarch does not vote, and neither do members of the immediate family. His or her main task is to represent the state because none of the other organs of the state can do that due to the principle of partition. It follows naturally that the monarch is obligated to avoid public discussions of a political nature and to stay neutral in any political struggle or conflict. However, this obligation is not stated explicitly in the Constitution.

History The foundation for the present monarchy was laid in 1849 when Denmark, as one of the last countries in Europe, acquired a liberal constitution. It happened during a civil war fought over the independence of a united Schleswig-Holstein. Denmark was a composite, colonial state and an absolute monarchy, codified by the Lex Regia of 1665. The king was given all powers, but at the same time, he was bound by law, so his rule was absolute but not arbitrary. The government faced several challenges in the 1840s. The Oldenburg dynasty was dying out, and the union between Denmark proper and the Duchy of Holstein—a member of the German Federation—would be dissolved like the Hanoverian-British union a few years earlier if a solution to the succession crisis was not found. A constitutional reform had been halted for decades due to the composite nature of the monarchy. At the same time, nationalism as a political movement was on the rise. The conflicts converged in March 1848 when the new king, Frederik VII, gave in to pressure from the Copenhagen public to replace the government with representatives of the so-called national-liberal movement which was working towards a unification of Schleswig and Denmark proper under one constitution. This in turn triggered an uprising by supporters of a unified Schleswig-Holstein, strongly supported by public opinion in most German countries. A constitutional assembly was formed through a combination of royal appointments and a general election in October 1848, and the assembly passed a constitution that was signed and ratified by the King on 5 June 1849. The Constitution did not replace Lex Regia completely; two articles concerning the house rules of the royal family are still in effect today. Known as Danmarks Riges Grundlov, literally ‘The Founding Law of the Realm of Denmark’, the Constitution had an ambiguous nature because it had strong elements of popular sovereignty, and at the same time, it was promulgated by the King and was clearly styled as such. Despite the efforts to create a unified Schleswig-Danish constitutional monarchy, the civil war ended with the status quo ante intact, and the Constitution was valid only for Denmark proper including the Faroe Islands, whereas the duchies—and Iceland—remained under direct monarchical rule.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

48   Jes Fabricius Møller At an international conference in London in 1852, the crisis of succession was solved with the election of Prince Christian of Schleswig-Holstein-Sonderburg-Glücksburg and his wife Princess Louise of Hesse-Kassel as heirs to the Danish throne, but neither the national nor the constitutional problem could be solved during the following decade. An attempt by the Danish government to pass a common constitution for Schleswig and Denmark proper in November 1863 triggered a declaration of war from the German Federation that was executed mainly by Prussian and Austrian troops. The Danish government probably had hoped that the other great powers would intervene and limit the conflict, thus providing a solution to problems that had proven politically insoluble. However, due to unfortunate circumstances, the unforeseen brilliance of Chancellor Bismarck, and a fumbled diplomacy, all three duchies Lauenburg, Holstein, and Schleswig were ceded to the enemy, and eventually, absorbed by Prussia like Hanover. Since 1849, the Constitution has defined the form of government as limited or restricted monarchical, and the king actually appointed the governments throughout the second half of the nineteenth century. In 1872, the liberal opposition, under the label ‘the Liberal Party’ (Venstre) won an absolute majority in the second chamber of Parliament (Folketinget). The Liberals’ main demand was that government formation should be based on the majority in Folketinget because it was elected by a broader elect­or­ate than the Upper House (Landstinget). However, King Christian IX (reg. 1863–1906) nevertheless kept appointing representatives of the landed elite as heads of conservative governments, provoking a constitutional crisis that was not solved until 1901 when the Liberal Party was eventually invited to form the government. As a rule, no government since has been able to function with a majority in the Folketinget against it. This did not mean that the king ceased to have legitimate influence over government formation. In 1920, King Christian X made the mistake of sacking the government without making sure that there was a majority against it and even without the counter-signature of the prime minister himself. This was the last time a king acted on his own in political matters. It caused a severe constitutional crisis known as the Easter Crisis (Påskekrisen). Demonstrators in the street demanded the King’s abdication and a republican constitution. Political leaders, in particular Social Democrats, negotiated a solution to the crisis and ended up supporting the monarchy. During the following years, a long-lasting and stable relationship between government and the head of state was established. For a short while during the German occupation from 1940–5, the King played an active, almost oppositional role in politics even though he stayed loyal to any government decision. Since 1920, the rule of ministerial counter-signature has ensured that the monarch never acts on his own in matters of state.

Succession Denmark has been a hereditary monarchy since 1660. According to Lex Regia, succession to the throne was agnatic-cognatic. In the eighteenth and early nineteenth century, it

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

The Monarch: Head of State and National Symbol   49 was believed in Copenhagen that Lex Regia was also valid in Schleswig but that was contested, particularly by the Schleswig-Holstein independence movement that claimed that the two duchies both followed the purely male succession of most German mon­ arch­ies. As part of the international agreements after the First Schleswig War 1848–50, succession was equalized in the entire Danish monarchy to become purely agnatic. The Act of Succession was revised in 1953 to give women restricted access to the throne and again in 2009 to give women equal access according to the principle of primogeniture. The heir to the throne is considered king or queen regnant the moment their predecessor dies. Neither abdication nor usurpation has taken place since the sixteenth century. Anointment or crowning has not taken place since King Christian VIII (reg. 1839–48) ascended the throne. Neither the throne nor the regalia are in use either, and speaking of the throne or the crown is purely metaphorical. The most important ceremony when a new monarch takes up office is the proclamation performed by the prime minister from the balcony of Christiansborg Palace, which houses the Supreme Court, the Parliament, and the offices of the prime minister. The monarch must swear loyalty to the Constitution before taking office (Section 8). Usually, this oath is taken when the heir to the throne joins the Council of State (Statsrådet) at the age of 18. The monarch is also obliged to belong to an ‘Evangelical Lutheran Church’, which, in theory, could be another church than the established Church of Denmark ( folkekirken), but it is con­ sidered self-evident that the monarch and the heirs to the throne are members of the established church. Other members of the royal family have freedom of religion. It is customary, however, that people from other denominations convert when marrying into the royal family. Situations in which the monarch is unable to govern due to illness or absence are regu­lated by law (1871). If the monarch is out of the country for more than a day— officially or privately—a governor of the realm (Rigsforstander) is appointed so that the country is not without an acting head of state. According to the law, any person of age who is not previously convicted and is a Danish citizen of Evangelical Lutheran faith can be appointed Governor of the Realm. As a rule, members of the royal family are chosen. When the heir apparent is so appointed, his title is Regent.

The Royal House The civil list—the size of the yearly allowance and the number of buildings that are at the monarch’s disposition—is decided by law with reference to Section 10 of the Constitution at the beginning of the monarch’s reign. This amount is then index regulated. In 2017, the annuity for the Queen was approx. DKK 83 million. The court has a staff of 110 people, which equals 95 full-time employees (Royal House 2017). Not covered by the civil list are guards, escorts, and military parade corps or transportation on the royal yacht, which is a vessel of the navy, air force planes, and helicopters. The total public expenditure for the monarch is the object of regular press scrutiny. The royal palaces are protected by the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

50   Jes Fabricius Møller military, and personal security of royal family members is taken care of by the Danish Security and Intelligence Service (Politiets Efterretningstjeneste). The cost of these services and the size of the staff are not published. According to Section 11 of the Constitution, annuities for individual members of the royal house are allowed. ‘The royal house’ is not defined by law, and allowances have not followed any general rule. Both members of the immediate and extended family, within and outside of the line of succession, have received allowances. Currently, the Queen’s cousin Count Ingolf of Rosenborg, who was the heir apparent until 1953, and both her sons receive an allowance. In the next generation, it is foreseen that only the eldest child of the Crown Prince will receive an allowance. Section 21 of Lex Regia states that no ‘prince of the blood’ may marry without the consent of the monarch. This is specified in the Law of Succession. Consent is given in the Council of State (Act of Succession, Section 5), that is, by the government and the monarch collectively. An heir to the throne who marries without royal consent may lose the rights to succession and the royal title. This provision has been applied several times in the twentieth century. Section 25 of Lex Regia states that ‘princes and princesses of the blood’ should not answer to the regular courts, only to the monarch. This is usually referred to as legal immunity, but it does not mean that members of the royal house cannot be punished. The provision dates back to when the king was actually head of the Supreme Court. The monarch herself enjoys full legal immunity (Section 13 of the Constitution). The crim­inal code does not contain articles on lèse-majesté. However, it is an aggravation if certain crimes are committed against the king or queen (Sections 112, 113, and 115).

The Court The royal court as an institution is not defined by law. It consists of the employees who serve in the monarch’s household and in the official functions of state. The court is financed by the civil list. Terms of employment equal those of public employees in general. The court is headed by the monarch. The two top officials are the Cabinet Secretary and the Lord Chamberlain. Their functions are described by the court itself as such: The Lord Chamberlain’s office is the secretariat for The Queen. The office is in charge of all the arrangements for official functions, such as state visits at home and abroad, dinners and luncheons, and court ceremonies, including presentation of credentials by ambassadors as well as their farewell audiences. [. . .] The Cabinet Secretary to HM The Queen heads the Cabinet Secretariat and is the link between HM The Queen and the Government, particularly the Prime Minister’s office and the Ministry of Foreign Affairs. He is the advisor and executive officer to HM The Queen in regard to her attendance at official engagements, her patronage, reception of deputations, audiences, matters pertaining to court titles and rank, and drafts of Her Majesty’s speeches. The Cabinet Secretary to HM The Queen also serves as secretary to the

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

The Monarch: Head of State and National Symbol   51 Chancery of the Royal Danish Orders of Chivalry, and in that cap­acity advises The Queen in regard to the awarding of orders and medals.  (Royal House 2019)

The Chapel of the Royal Orders of Knighthood The Chapel of Chivalry (Ordenskapitlet) administers the two awarded orders: the Order of the Elephant and the Order of Dannebrog. The latter is divided into six classes: Grand Commander, Grand Cross, Commander of the First and Second Degree, and Knight of the First and Second Degree. The Chapel of Chivalry serves directly under the monarch as a personal prerogative, and orders are awarded without the countersignature of a minister. The cabinet secretary is also daily head of the Chapel of Chivalry and answers directly to the monarch. Candidates for decoration are nominated by the heads of ministerial departments. Politicians are not to exercise influence on decorations. The exchange of decorations during state visits is coordinated with the Ministry of Foreign Affairs. Holders of the two lowest classes of the Order of Dannebrog are called knights, but it is not an aristocratic title. Neither personal nor hereditary peerages have been awarded to commoners for more than a hundred years. On occasion, members of the royal family have been elevated into the ranks of nobility when their royal title has been taken away from them, as in the case of marriage without royal consent or divorce. When Prince Joachim’s first wife remarried in 2007, she was dubbed Countess of Frederiksborg. In 2008, both of the Queen’s sons were given additional titles as Counts of Monpezat, thus domesticating the Prince Consort’s family name as a Danish hereditary aristocratic title.

The Monarch’s Functions ‘The signature of the King to resolutions relating to legislation and government shall make such resolutions valid, provided that the signature of the King is accompanied by the signature or signatures of one or more Ministers’ (Constitution, Section 14). The Queen signs government proposals and legislation. It has occurred that a law has been without validity because it had not been signed by the Queen within the given time limit of 30 days (Constitution, Section 22), but that was due to an omission by the legislative branch. The monarch’s right of veto is purely theoretical. The government’s law proposals are presented in the Council of State, which convenes 5–10 times a year. The Council of State consists of all ministers of the government and the monarch, who chairs the council. When coming of age at 18, the heir apparent ‘has a

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

52   Jes Fabricius Møller seat’ on the Council of State (Constitution, Section 17). HM Queen Margrethe II has thus taken part in meetings on the Council of State for more than 60 years. The Queen attends the opening meeting in the Folketinget every year on the first Tuesday of October. Previous monarchs occasionally gave the opening speech stating the intentions of the government, the so-called ‘throne speech’, but the Constitution (Section 38) now directs the prime minister to give this speech (Nielsen 2012). The Queen has separate regular private audiences with the prime minister and the foreign minister with no other persons attending. The origin of these conversations or consultations goes back to the time when the king actually exercised influence on government affairs, but they are also valuable today because if the Queen is supposed to stay out of politics, she needs to be informed about what is on the political agenda. One of the main tasks of the monarch is to represent Denmark when receiving ambassadors and when receiving and paying official and state visits. On official visits, the monarch is always accompanied by one or more ministers who carry the political responsibility. Other members of the royal family can make visits or receive foreign guests, but the head of state is always received by a head of state on official visits. Decorations are exchanged by prior agreement (Møller 2014). There is clearly an increasing tendency for members of the royal family to take part in activities in support of Danish businesses when going abroad. Attempts have been made to measure the economic value of these visits, but they are not precise (Sørensen and Aksglæde 2018). Top civil servants and military officers are formally employed by the monarch, who signs the papers of employment and promotion without having any influence on the process except when it comes to employees of her court. For historic reasons, the Queen also approves new hymn-books and altar books of the established church. She is sometimes described as head of the church, which is not correct. The monarch’s role is in­herit­ed from the absolutist past since no legislation regulating the established church or defining its leadership has been passed since 1849. The church has no clear hierarchy or chain of command and is highly decentralized. As far as authority exists, it is exercised by the minister of church affairs. The monarch as an institution is of great importance to the relationship with the Faroe Islands, Greenland, and the Community (or Commonwealth) of the Realm (Rigsfællesskabet) (see Gad  2020). The North Atlantic possessions were Norwegian when they came under the Danish Crown in 1380 along with Norway. Until 1721, there had been no contact with Greenland for centuries. After the Treaty of Kiel in 1814, Norway was ceded to Sweden, while Greenland, Iceland, and the Faroe Islands remained under the Danish Crown. In 1918, the sovereign Kingdom of Iceland was created in personal union with Denmark. In 1944, Iceland became a republic with overwhelming popu­lar support. Since the Second World War, the Faroe Islands and later Greenland have gained a wide degree of autonomy in all matters other than foreign policy, defence, and justice. The situation today increasingly resembles the relationship with Iceland after 1918. Royal visits to Greenland and the Faroe Islands are relatively frequent, and the Queen never fails to mention both in her televised New Year’s Eve address.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

The Monarch: Head of State and National Symbol   53

The Legitimacy of the Monarchy Two main arguments are used to criticize the Danish monarchy and/or advocate for the dissolution of the monarchy. Firstly, the monarchy is expensive, and it is inappropriate for taxpayers to promote social inequality and finance royal ‘lives of luxury’. Generous gifts to the royal family from businesses and other sponsors have similarly been objects of criticism (Høvsgaard 2012). Secondly, monarchical privilege based on birth defies the basic principles of democracy and meritocracy. As a recent textbook for students of law states: ‘The idea that the office as head of state should be inherited by a member of a certain family is obviously out of sync with contemporary views on egalitarianism and democracy’ (Christensen et al. 2016: 49). The republican movement in Denmark, such as it is, is neither united nor strong. Only one party currently represented in Parliament, the Red-Green Alliance, is actively against the monarchy, but it does not push the agenda. Important representatives of other parties (the Alternative and the Socialist People’s Party) have expressed concerns and apprehensions about some monarchical institutions. Historically, both the Social Liberals and the Social Democrats have had the abolition of the monarchy as an official policy but not any longer. However, a significant minority in both parties do still hold that opinion. More generally, members of Parliament and members of government belonging to either party, along with the Socialist People’s Party and the Red-Green Alliance, do not accept decorations. Arguments in favour of the monarchy—or legitimation strategies—can be cat­egor­ ized into four groups. First comes the formal or legalistic argument that the monarchy is given by the Constitution. The difficulty of changing the Constitution (see Christensen 2020) provides both de jure and de facto safeguards. As mentioned earlier, difficulties arise because the wording of the Constitution is a far cry from reality in many cases. The legal argument is related to the second argument, the democratic or popular. Since the present Constitution and Law of Succession were confirmed by a plebiscite in 1953, the monarchy can claim a considerable degree of democratic legitimacy. The Law of Succession was revised in 2009 and also confirmed by a plebiscite. Opinion polls since the late 1960s have shown that the monarchy and the monarch enjoy a high degree of popular support. Depending on the questions asked, 70-90 per cent approve of the monarchy or the monarch. Not surprisingly, political conservatives show more support for the monarchy, but left-wing support is also strong. When asked in 2013 if they agreed or disagreed with the abolition of the monarchy, 94 per cent of political conservatives disagreed, while the corresponding figure for members of the Red-Green Alliance was 66 per cent. Support or lack of support is not specific to generations: 82 per cent of 18–35-year-olds and 84 per cent of the 60+ age group disagreed (TNS Gallup 2013). The third argument is functionalist or pragmatic. The justification of the monarchy is that it works, or at least that it does not have a negative impact on society as such. In support of this view, it is noticeable that very few members of Parliament and no present or

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

54   Jes Fabricius Møller former prime minister or ministers have spoken out against the monarchy. There have been no serious attempts to change the state form since 1920. Conversely, this functionalist approach to the monarchy also puts pressure on the monarch always to act in accordance with the expectations of politicians and the public. Finally, the monarchy has the legitimacy of tradition. The monarchy represents stability and continuity, both of which are considered important in times of rapid change or turmoil. This is most often an expression of a more pragmatic than ideological approach. Supporters of the monarchy rarely define themselves as monarchists or anti-republican for that matter. Age and tradition are often emphasized in a more general sense, i.e. that Denmark is one of the oldest monarchies in the world (Cannadine 1983).

Conclusion The role of the monarch as head of state is comparable to that of presidents and monarchs alike in the parliamentary democracies of Northern Europe. This raises the question of how this purely ceremonial and symbolic role is to be understood. There is no simple answer to that question. One element to be aware of is that ceremonies and symbols matter. Even though HM Queen Margrethe II has no formal political authority, she is still the highest-ranking person in Denmark and attracts attention wherever she goes. The monarch and the royal house are the centre of a symbolic order that can be viewed as both egalitarian and hierarchical. On the one hand, everybody is equal in being symbolically subordinate to the monarch, but on the other, there is a clear and visible hierarchy. It has been attempted to measure citizens’ place in this hierarchy by proximity to the royal house, and it gives prestige to be invited to official events. An attempt of assessing who is most influential in Denmark includes a place on the royal guest list as a relevant factor. Using Bourdieu’s theory, Larsen et al. (2015: 28) thus find that businesspeople in particular are able to convert economic capital into social and (cultural) capital due to their contacts in the royal family. However, the survey does not clarify the direction of causality. Does proximity to the court generate influence, or do influential people tend to seek proximity to the court? Twentieth-century Danish monarchs have gradually adapted to an increasingly egalitarian society. They have reached out to the general public through the media, by systematically visiting every corner of the realm, and by awarding medals of merit to loyal private and public employees with a tenure of more than 40 years. Recipients of medals and orders (1,942 medals of merit and 455 Orders of Dannebrog in 2018) are invited to public audiences. The chance that a Dane in the course of a lifetime meets a member of the royal family is not small. But the public image of the monarch and the royal family is primarily maintained by a careful media strategy, taken care of by a press secretariat in the Lord Chamberlain’s office since 2004. And, as the opinion polls indicate, the strategy has been successful.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/13/2020, SPi

The Monarch: Head of State and National Symbol   55 The monarch is thus not only a representative of the state but also—and perhaps more importantly—a symbol of or for the nation. This makes the monarchy as an institution dependent on public affection for the reigning monarch. The popularity of the reigning Queen is confirmed annually on her birthday, 16 April. At precisely noon, she enters the balcony of her residential palace in Copenhagen to be greeted by a crowd of thousands cheering and waving flags. When the Crown Prince turned 50 in 2018, it was announced that he would do the same on his birthday, 26 May. The palace square was even more crammed with cheering people than the month before. This was generally interpreted as an assurance that the Danish monarchy will maintain its status in the following gen­er­ ation as well.

References Andersen, Poul (1954). Dansk Statsforfatningsret. Copenhagen: Gyldendal. Cannadine, David (1983). ‘The context, performance and meaning of ritual’, in Eric J. Hobsbawm and Terence O. Ranger, eds, The Invention of Tradition. Cambridge: Cambridge University Press, 101–64. Christensen, Jens Peter (2020). ‘The constitution’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 9–27. Christensen, Jens Peter, Jørgen A.  Jensen, and Michael  H.  Jensen (2015). Grundloven med kommentarer. Copenhagen: Jurist- og Økonomforbundets Forlag. Christensen, Jens Peter, Jørgen A.  Jensen, and Michael  H.  Jensen (2016). Dansk Statsret. Copenhagen: Jurist- og Økonomforbundets Forlag. Gad, Ulrik P. (2020). ‘Greenland, the Faroe Islands, and Denmark. Unity or community?’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 28–45. Høvsgaard, Jens (2012). Det koster et kongerige. Copenhagen: Rosinante. Larsen, Anders  G., Christoph Ellersgaard, and Markus Bernsen (2015). Magteliten. Copenhagen: Politikens Forlag. Møller, Jes F. (2014). ‘Monarkiet som Danmarks ansigt udadtil’, Økonomi og politik, 87/4: 37–47. Nielsen, Peter  H. (2012). ‘Statsministeren fik ordet. Historien om den danske åbningstale’, Økonomi og Politik, 85/4: 87–99. Royal House (2017). ‘Årsrapport fra Kongehuset 2017’, http://kongehuset.dk/aarsrapport-2017/ oekonomi Royal House (2019). ‘The Court’, http://kongehuset.dk/en/the-court (accessed 15 May 2019). Sørensen, Jonas, and Jacob Aksglæde (2018). Kongehuset og diplomatiet. Copenhagen: Forlaget Politiske Studier. Sørensen, Max (1973). Statsforfatningsret, 2nd ed. by Peter Germer. Copenhagen: Jurist- og Økonomforbundets Forlag. TNS Gallup (2013). Gallup om kongehuset, 16 December 2016.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

chapter 5

The Elector a l System Fair and Well-Functioning Jørgen Elklit

Arend Lijphart, the doyen of electoral systems research, has expressed his appreciation of the system used to elect the Danish Parliament (Folketinget) on several occasions. One example is this quote: Of the electoral systems analyzed in this book, I nominate the Danish system as the closest approximation to my ‘ideal’ model. [. . .] [L]et me merely highlight its main features here: list PR, an average district magnitude of about eight seats, national compensatory seats with a low 2 per cent threshold, and highly proportional allocation formulas. My one misgiving concerns the high degree of openness of the list system and the complexity of how the partly open lists work. No system is perfect! [. . .] Those who designed the system almost a century ago did a much better job than their contemporary counterparts [. . .]  (Lijphart 2005: ix)

The quote highlights some of the system’s main features, which are key to understanding how it can simultaneously be both complex and simple as far as Danish voters are concerned, regardless of whether they just want to cast a vote in the elections or optimize their actual electoral power. The system has now reached its centenary. It has developed incrementally since 1920 when it was negotiated and enacted in cooperation between the then Minister of the Interior Ove Rode and Conservative opposition MP Asger Karstensen. In 1915, Denmark had already introduced elements of proportional representation (PR) as single-member constituencies were combined with a system of compensatory seats outside the capital district of Copenhagen and Frederiksberg, which was made into a multi-member PR district (Elklit 1992; Shugart and Wattenberg 2001: 580). However, for a number of reasons, the system did not achieve an acceptable level of proportionality in the 1918 Folketinget elections.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   57 It was consequently changed fundamentally in 1920 as one of several elements in the political settlement after a major constitutional crisis. This new electoral system was, in many respects, quite similar to the current system. It is indeed remarkable that the basic features of the system have survived for 100 years. The main reason is that incremental changes over this period have allowed the electoral system to adjust to social, demographic, and political changes. Folketinget elections are generally perceived to be of a high standard when it comes to election administration and the general quality of the election processes. Other members of this top league are Finland, Norway, Iceland, Germany, the Netherlands, and Sweden; it is interesting that these countries all have PR electoral systems even though their systems vary. The quality league scoring is documented by the Electoral Integrity Project (Norris et al. 2018). It also deserves mention that Denmark adheres to the ‘governmental model of electoral management’ (Catt et al. 2014) as elections are organized and managed by the Ministry of Social Affairs and the Interior in close cooperation with local authorities. An English language text of the electoral law is available at the Ministry’s home page (cf. below the references). Furthermore, an annotated edition of the electoral law was recently published (Miller and Elklit 2019). This chapter deals only with the parliamentary electoral system, and it follows to a considerable degree the conceptualizations and notions presented in Herron et al. (2018). The electoral system used in regional and municipal elections (and European Parliament elections) is a straightforward list PR d’Hondt system allowing for apparentement, something not possible in Folketinget elections (Houlberg and Ejersbo 2020; Kjær 2020; Elklit 2016a: 48–55). Election results are accepted by all, the electoral administration (the electoral management body, EMB), based in a small office in the Ministry, functions well and is never accused of any kind of political bias, and general trust in the system is widespread. One likely explanation for this positive state of affairs is that Denmark has not had a majority party in Parliament since 1909. All political parties have to realize that they must compromise to get legislation passed and that small parties might be necessary coalition partners. Political parties also appreciate that the electoral winners of today may be the losers of tomorrow, so the losers of today should not be treated too harshly. The structure of the chapter is as follows. The first section presents the electoral ­system as it has developed since 1920. The second section describes in more detail how the system works, with all descriptions and explanations based on the outcome of the Folketinget elections on 5 June 2019. The third section looks at a couple of electoral law amendments to illustrate how such changes have been discussed and enacted, the purpose being to see if the political processes leading to the amendments reflect the fact that electoral laws are of a special, consensus-driven nature in Denmark. The fourth section presents a few comparisons with electoral systems elsewhere. A brief conclusion closes the chapter.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

58   Jørgen Elklit

The Structure of the Electoral System The electoral system is a straightforward, two-tier PR system where the national (upper) level is the decisive level (cf. Lijphart 1994: 32–6; Elklit and Roberts 1996), while lower level seat allocation takes place in multi-member constituencies (MMCs). This ensures that MPs have some kind of link to their constituencies. This is even more so in relation to the nomination districts (opstillingskredse), a sub-division of the MMCs where the local party branches nominate their candidates to stand for election. Independent candidates are allowed, but only very few stand for election; they are also linked to a nom­in­ ation district even though the relevant constituency for both kinds of candidates is the MMC. The basic features of the system are described in Table 5.1, while Table 5.2 provides an overview of the development over time in the system’s key features. Further details will be provided in the subsequent section where the system’s actual workings are described. Table 5.2 shows that the system has gone through seven phases since its inception in 1920. However, the picture softens if we look at the nature of the changes involved. For a general discussion of electoral system changes, see Renwick (2018). In 1920, after the passing of the electoral law mentioned above, North Schleswig was reunited with Denmark following plebiscites in accordance with the post-First World War Versailles Settlement. The electoral system consequences were an additional MMC, North Schleswig, and a slight increase in the number of seats in the Folketinget. Lijphart (1994: 13) has suggested that one should only speak of electoral system changes in two-tier systems either if the electoral formula at the decisive level is changed or if there is a change of 20 per cent or more in at least one of three important electoral systems features, namely the district size in the decisive tier, the size of the formal elect­ or­al threshold, and the assembly size. The upper-level seat allocation formula has been the same since 1920, and the changes in the district size at the decisive (national) level—and the assembly size by implication—have not surpassed the 20 per cent criterion on any occasion. Table 5.3 shows that it is only electoral threshold changes that have generated electoral system change. The first of three changes took place in 1948, the second in connection with the constitutional amendments in 1953, and the third in 1961. Consequently, one can say that the current manifestation of the Danish electoral system comprises all twenty-one Folketinget elections since 1964. In 2005, a local government reform led to a substantial reduction in the number of municipalities as well as the abolition of the counties which were replaced by five administrative regions. The counties had been used as MMCs since 1920, but having only five MMCs was not considered a good idea, so the number of MMCs was only reduced from seventeen to ten. A consequence was an increase in the average number of MMC constituency seats which is now 13.5 (i.e. considerably more than Lijphart saw as praiseworthy in the above quote). This might have entailed easier access to seats in Parliament

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   59

Table 5.1.  The Overall Structure of the Danish Electoral System Level (tier)

Central Features

Effects/Consequences

Seats are allocated using Hare + Largest Remainders (LR) for parties surpassing at least one of three electoral thresholds (the most important being 2 per cent of the valid vote).

High level of overall proportionality at decisive level (see Table 5.3)

Of the 175 seats allocated in Denmark (i.e. excluding the two seats in Greenland and the two given to the Faroe Islands), 40 seats, i.e. 23 per cent, are compensatory seats.

Instrumental in reaching the high level of proportionality (cf. Taagepera and Shugart 1989: 131)

Lower (multi-member constituencies [MMCs] most often fitted to other administrative borders)

The number of seats in each multi-member constituency (MMC) is determined for a five year period based on the sum of the total population, the size of the electorate, and a factor reflecting the geographical space of the MMC in question. Seats are allocated within the MMC using a PR allocation formula. Since 2005, it has (again) been the d’Hondt formula.

Allocation of seats across the country is considered fair and reasonable. Likewise, the system of allocation of seats within the MMCs is seen as unproblematic, mainly because the allocation of compensatory seats makes up for some of the biases created by the d’Hondt formula.

Eventual (intra-party) seat allocation to individual party candidates

The MMC branches of the political parties decide on which of several available systems they will use for nominating and ordering their candidates in the MMC. This includes how the seats won eventually will be allocated to the party’s candidates.

When parties organize their ‘list’ of candidates in an MMC, they must choose between different systems with different consequences for the eventual selection of the individual candidates. This unique feature explains to some degree the complexity of the Danish electoral system (cf. Lijphart’s complaint in the quote above). It allows MMC party branches to decide which form of intra-party competition they want.

Voters have one vote which is cast either for a party or for a specific candidate (i.e. a preferential or a personal vote). Lists are either open or various kinds of semi-open, never completely closed. In Folketinget elections, about 50 per cent of the voters cast a preferential vote (52 per cent in 2019).

The possibility of running on completely open lists was first available in 2019. It was used in 46 per cent of all 130 possible cases (13 parties in 10 MMCs).

National

Table 5.2.  The Development in the Main Features of the Electoral System, 1920–2020 Period (by elections)

1920, April–1920, July

1920, Sept.–1947

1950–1953, 1953, Sept.–1960 April

1964–1968

1971–2005

2007–2019

2

10

2

3

3

14

4

Hare + LR

Ditto

Ditto

Ditto

Ditto

Ditto

Ditto

Total number of seats1

139

148

149

175

Ditto

Ditto

Ditto

Electoral thresholds

One constituency seat Ditto or as many votes as the national votes/ seats average in one of the three regions;2 eff. electoral threshold approx.: 1.1 per cent

Ditto

One constituency seat or 60,000 votes or as many votes as the regional votes/constituency seat average in each of the three regions;2 eff. electoral threshold approx.: 2.6 per cent

One constituency seat or 2.0 per cent of all valid votes or as many votes as the regional average pr. constituency seats in two of the three regions;2 eff. electoral threshold: 2.0 per cent

Ditto

Ditto, but the three regions are delineated differently

Compensatory seats

29

31

44

40

Ditto

Ditto

Ditto

Compensatory seats in per cent of total

21

Ditto

30

23

Ditto

Ditto

Ditto

Number of MMCs

22

23

Ditto

Ditto

Ditto

17

10

Periodical adjustment of allocation of seats to MMCs

No

Ditto

Yes

Ditto

Ditto

Ditto

Ditto

Average number of constituency seats in MMCs

5.0

5.1

4.6

5.9

Ditto

7.9

13.5

MMC seat allocation formula

d’Hondt

Ditto

Ditto

Modified Sainte- Laguë

Ditto

Ditto

d’Hondt

Allocation of constituency seats final?

Yes

Ditto

No

Yes

Ditto

Ditto

Ditto

1 Excluding the Faroe Islands and Greenland. 2 The three regions were the capital region (Copenhagen and Frederiksberg municipalities); the islands; and Jutland. This regionalization remained in use until the changes in regional delimitation in the wake of the 2005 Local Government Reform.

OUPOUP UNCORRECTED CORRECTEDAUTOPAGE AUTOPAGEPROOFS PROOFS––, FINAL, 06/15/2020, 06/15/2020, SPiSPi

Number of elections Upper level allocation formula

Period (by elections)

1920, April and June

1920, Sept.–1947

1950–1953, April

1953, Sept.–1960

1964–1968

1971–2005

2007–2019

Number of elections

2

10

2

3

3

14

4

Upper (decisive) level allocation formula

Hare + LR

Ditto

Ditto

Ditto

Ditto

Ditto

Ditto

Change in district size (decisive level)

NA

+ 7%

+1%

+ 17%

NA

NA

NA

Change in electoral threshold

NA

NA

+ 45%

+ ca. 73%

− ca. 23%

NA

NA

Change in assembly size

NA

+ 7%

+ 1%

+ 17%

NA

NA

NA

Gallagher’s index of disproportionality

0.01

0.02

0.01

0.02

0.02

0.02

0.01

Effective number of electoral parties

3.8

3.7

4.0

3.8

4.2

5.3

5.9

Effective number of parliamentary parties

3.7

3.6

3.9

3.7

3.9

5.0

5.7

r

0.03

0.02

0.01

0.05

0.06

0.05

0.04

Note: Figures are calculated with more decimals than shown and then rounded to increase readability.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Table 5.3.  Changes in Lijphart’s Four System Change Indicators and Average Values for Gallagher Disproportionality Scores, Effective Number of Parties (N(v) and N(s)), and r

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

62   Jørgen Elklit for small parties since the natural electoral threshold decreases when M, the number of seats in a constituency, increases. However, this possibility made some fear that small parties might win a ‘cheap’ constituency seat in a large MMC, giving such parties access to the Folketinget without having won even 2 per cent of the national vote total. Consequently, the Modified SainteLaguë formula used in the MMCs since 1953 (Elklit 1999) was replaced by the d’Hondt formula which does not provide ‘cheap’ seats to small parties. So far, this safeguard aimed at avoiding unintended consequences stemming from the decrease in the number of MMCs has worked as expected. Therefore, the changes subsequent to the local government reform are best understood as an ordinary adjustment to changes in other administrative spheres and not as an electoral system change per se. Table 5.3 also provides the information necessary to assess if the changes in electoral system features identified above have been followed by systematic changes in disproportionality scores (Gallagher 1991, 2005) or in the value of r, the relative difference between the effective number of electoral and parliamentary parties (Laakso and Taagepera 1979; Taagepera and Shugart 1989: 204, 209). Figure 5.1 illustrates the development of the effective numbers of parties since 1918. The level of disproportionality, as well as 8 7 6 5 4 3 2

0

1918 1920–04 1920–07 1920–09 1924 1926 1929 1932 1935 1939 1943 1945 1947 1950 1953–04 1953–09 1957 1960 1964 1966 1968 1971 1973 1975 1977 1979 1981 1984 1987 1988 1990 1994 1998 2001 2005 2007 2011 2015 2019

1

N(v)

N(s)

Figure 5.1 Effective number of parties in the electorate, N(v), and in Parliament, N(s), 1918–2019.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   63 the relative distance between N(v) and N(s), that is r, is remarkably low—also by ­international comparison—which explains why the two lines are so close to each other. The disproportionality score (and r) nevertheless increased after 1953 when the Constitution was amended and the electoral law was adjusted to reflect these amendments. The changes in the electoral threshold in 1948, 1953, and 1961 did not cause changes in the two effect variables considered here. The reason behind the increase in the r value during the 1960s, clearly visible in Figure 5.1, is the party system development between the mid-1960s and the mid-1990s where differences in party size—in the electorate and in the Folketinget—decreased the same time that some of the parties did not meet the 2 per cent threshold. The number of substantial changes in key electoral system features is thus reduced to three (as indicated by the bold vertical lines in Table 5.3), a dramatic change after the 1947 elections; the amendment of the Constitution in 1953, which also entailed an amendment of the electoral law in 1953; and then an amendment in 1961 of the rules related to the electoral thresholds. The latter is discussed below as it illustrates how small parties can also sometimes have it their way.

How the Electoral System Actually Works Political parties can participate in elections to the Folketinget if they fulfil one of two conditions: (1) they won representation in the previous Folketinget elections and still are represented in Parliament, or (2) as a new party that has registered at the Ministry of Social Affairs and the Interior and has submitted a number of supporting signatures from registered voters corresponding to at least 1/175 of the number of valid votes in the previous elections (the average ‘price’ of one seat). This number was 20,109 in 2019, and three new parties registered in this way. For the next elections, the corresponding number is 20,182. Ten parties fulfilled the first criterion in 2019. A digital system for the handling of supporting signatures was implemented in 2016. The procedure is one of two steps with a period of pensiveness in between to ensure that the voter is positive that she wants to submit her signature. Two of the three new parties were in a hurry to submit the signatures, and they realized that a shortcut was possible so that the signatures could be submitted faster than prescribed by the Ministry. This created some discussion, but the Ministry decided to accept the signatures, promising that the shortcut would be closed as soon as possible. The number of required supporting signatures for new parties is considerably higher than in, for instance, Sweden and Germany, and it is seen by some as an additional electoral threshold. Inter-party and intra-party seat allocation takes place in six sequential steps. Taken together, these steps reflect the intentions behind the electoral law as it has developed over the years: overall proportionality for parties with a minimal level of voter support, regional

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

64   Jørgen Elklit scattering of seats, locally rooted candidates, and voter influence not only on the political party composition of the Folketinget but also on the selection of the individual MPs. The two seats allocated to Greenland and the two allocated to the Faroe Islands are allocated separately, using d’Hondt in both cases. The consequence is that the two biggest parties in each of these two parts of the realm normally win a seat each. Figure 5.2 illustrates the entire allocation process where one finds the allocation of the four North Atlantic seats just below the upper left corner. The 175 seats available for the third part of the realm (Denmark) are the sum of the 135 constituency seats allocated directly in the MMCs in the first of the six steps in the seat allocation process and the 40 compensatory seats.

Step 1 The 135 constituency seats are allocated proportionally for five-year periods to the ten MMCs based on a formula that includes population size, the size of the electorate,

179 seats

Greenland: 2 seats (d’Hondt); Faroe Islands: 2 seats (d’Hondt)

175 seats = 135 MMC seats + 40 compensatory seats

Step 1: 135 MMC seats (by d’Hondt)

Step 6: Finding out which individual party candidates in the MMCs will have the constituency and the compensatory seats won. This depends on how the parties themselves have determined what system shall be used

Step 2: 3 electoral thresholds

Step 3: Each eligible party’s total seat allocation is calculated using Hare + LR. A party’s number of compensatory seats is subsequently calculated by subtracting its total number of MMC seats from its total seat allocation. The resulting figure is the party’s number of compensatory seats

Step 4: Sainte-Laguë formula used to allocate compensatory seats won by parties to regions

Step 5: Compensatory seats won by parties in regions are allocated to MMCs using divisors 1, 4, 7, 10, etc.

Figure 5.2  The six steps of seat allocation in the Danish electoral system.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   65 and twenty times the MMC area in km2 (a consideration for sparsely populated parts of the country). The number of seats in the MMCs varied in 2019 between 2 (Bornholm’s MMC) and 20 (Zealand’s MMC). If these calculations do not result in Bornholm, an island in the Baltic Sea, getting at least two seats, the MMC is given two seats in advance, and calculations are redone for the other nine MMCs. This ensures that the island will have at least some proportionality in its political representation in the Folketinget, which is not possible with only one MP. The consideration given to geography is reminiscent of the small nineteenth-century constituencies in rural Jutland because of the peninsula’s low population density, particularly in its Western part. As a result, there is now a slight tendency for parties to win marginally more seats in Jutland. However, parties with strongholds in Jutland do not get more seats in total than they would otherwise have gained as the national tier is the decisive tier. Since the 2007 elections, the MMC constituency seats have (again) been allocated to political parties according to the d’Hondt seat allocation formula, and the Modified Sainte-Laguë formula—widely associated with Scandinavian elections—is no longer used in Denmark. This allocation is final, i.e. if a party should get more constituency seats than its overall seat entitlement, it can keep them all. The risk of that happening is, however, negligible.

Step 2 This step establishes which parties might be eligible for compensatory seats. A party must fulfil a least one of three requirements to qualify for inclusion in this allocation: (1) winning at least one constituency seat in any of the MMCs, (2) winning as many votes as the average of the regional votes/constituency seats in two of the three main electoral regions, or (3) winning at least 2 per cent of the national vote total. The latter is, for all intents and purposes, the effective electoral threshold. In 2019, where 13 parties had re­gis­tered for participation in the elections, 2 per cent was equal to 70,635 votes. Five parties did not win any constituency seats, but two of them, the New Right (Nye Borgerlige) and the Liberal Alliance (Liberal Alliance), both surpassed the 70,635 vote requirement and were thus entitled to participate in the sharing of the 40 compensatory seats.

Step 3 Each party’s overall seat entitlement is found by allocating all 175 seats to the eligible parties by using Hare + Largest Remainders (see Table 5.4). For instance, the Red-Green Alliance (Enhedslisten) was entitled to 13 of the 175 seats. They had, however, won 7 MMC seats, so they were entitled to 6 compensatory seats.

Votes

Total A. Social Democrats B. Social Liberals C. Conservatives D. New Right F. Socialist People’s Party I. Liberal Alliance O. Danish People’s Party V. The Liberals Ø. Red-Green Alliance Å. The Alternative Subtotal elegible parties E. Klaus Riskær Pedersen K. Christian Democrats P. Taut Course 13 Independent candidates

    914,882 304,714 233,865 83,201 272,304 82,270 308,513 826,161 245,100 104,278 3,375,288 29,600 60,944 63,114 2,774

175 seats allocated proportionally based on total vote figures

175 seats allocated proportionally based on total vote figures, rounded

Constituency seats won in MMCs

175.000 47.434 15.799 12.125 4.314 14.118 4.265 15.996 42.834 12.708 5.407

175 48 16 12 4 14 4 16 43 13 5

135 44 12 9 – 12 – 11 39 7 1

NA NA NA NA

NA NA NA NA

– – – –

Source: Statistics Denmark 2019: 4. The percentage distributions of votes and seats are added here.

Compensatory seats

40 4 4 3 4 2 4 5 4 6 4 NA NA NA NA

Per cent of total Per cent of all valid vote 175 seats

100.0 25.9 8.6 6.6 2.4 7.7 2.3 8.7 23.4 7.0 3.0 0.8 1.7 1.8 0.1

100.0 27.4 9.1 6.9 2.3 8.0 2.3 9.1 24.6 7.4 2.9 − − − −

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

Table 5.4.  The Official Result of the 2019 Folketinget Elections and the Computation of Compensatory Seats

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   67

Steps 4 and 5 These steps determine in which MMCs the parties will have their compensatory seats. Step 4 allocates each party’s entitlement to one of the three main electoral regions (to ensure geographical spread). The pure Sainte-Laguë formula is used across parties and regions, and the 40 highest quotients entitle the parties in question to a compensatory seat in the region in question, provided that the party has not yet reached its full entitlement and that the region has not transgressed its previously decided share of the compensatory seats (Statistics Denmark 2019: 15–16). In Step 5, parties have the compensatory seats allocated to regions in Step 4 further allocated to specific MMCs within the regions. The formula used is the so-called ‘Danish’ string of divisors (1, 4, 7, 10 . . .) which ensures geographical spread within the regions (because of the distance between the divisors) so that MMCs where a party is less strong also have a chance of getting a compensatory seat (Statistics Denmark 2019: 17–19). This feature was, however, more important prior to 2005 when MMCs were smaller.

Step 6 All 135 constituency seats and all 40 compensatory seats are now allocated to a particular party in a particular MMC. This final step is where it becomes clear which of each party’s candidates in an MMC will eventually become members of Parliament (MPs), namely, this is the process of intra-party seat allocation. At the time of registering the candidates prior to the election, political parties and their MMC branches (and local branches in the nomination districts) have a range of options. Denmark is unique in allowing pol­it­ ical parties the liberty of choosing between different ways of having their MPs selected from the pool (or list) of candidates. It is even the case in Denmark that a party can decide differently in different MMCs or even in different nomination districts within an MMC. The consequence is that the way voters cast their party or preference vote can have different effects when it comes to the eventual selection of the successful candidates. However, most voters probably do not understand the finer details of this part of the electoral system! MMCs are subdivided into a number of so-called nomination districts, the idea being that party branches at this level normally choose their own candidates for Parliament, while the party’s MMC branch (or the national party secretariat) will be responsible for registering the candidates with the registration authority. Some claim that the nom­in­ ation districts are not particularly important these days, which might be true. But today’s nomination districts are, to some degree, a continuation of the 1920 nomination districts which were the single-member constituencies used before that, and they are still a useful communication channel between voters and their representatives. In 1920, nomination and seat allocation was simple: candidates got the votes cast for their party in their own nomination district, as well as all preference votes cast for them in all nomination districts in the MMC, including, of course, their own. This was

OUPOUP UNCORRECTED CORRECTEDAUTOPAGE AUTOPAGEPROOFS PROOFS––, FINAL, 06/15/2020, 06/15/2020, SPiSPi

68   Jørgen Elklit pos­sible because all candidates for all parties in the MMC had their names on the ballot paper and could be voted for. Candidates for a party became MPs according to their vote sums in declining order. Those not elected became substitutes, also ranked according to vote sums in declining order. This nomination option is still a possibility, even though it is only used rarely; it is called ‘nomination by district’ (kredsvis opstilling), and it decides the order in which the party’s candidates are listed on the ballot paper. Each nomination district has its own distinct ballot paper, and for each party, the local candidate comes first (with the name printed in bold), followed by the other candidates in alphabetical order. Since 1920, various other ways of nomination and intra-party seat allocation have developed, mainly as a reflection of political and social developments (including decline in party membership, increased size of the MMCs, media structure, and changing administrative borders). Space does not allow for a full account of this development; elements thereof can, however, be found elsewhere (Elklit 2005/2008, 2016a). The most recent changes were enacted in 2017, seemingly as a concession from the government to the Danish People’s Party which was eager to obtain a couple of changes to the nomination and intra-party seat allocation system, inter alia to allow parties to list candidates on the ballot paper in the order of their own choosing. These changes took effect for the first time in 2019. Only time will show the effects of these procedures over time as they provide more options for parties than ever before. If an MMC has, say, nine nomination districts, most parties will have a candidate in each of the nine nomination districts. All nine candidates will also be on the ballot paper in all nine nomination districts and compete against each other (as well as against all the candidates from all the other parties) since all voters in an MMC can cast a preferential vote for any of the candidates. So even though a party has a list of candidates, they are primarily a pool of equal party candidates, all interested in winning. Their chances of success in the intra-party electoral game are not decided by their place on the party’s list of candidates but primarily by their party’s strength in their ‘own’ nomination district and the number of preferential votes they get in other nomination districts. An old saying in Danish electoral politics is that it is easier to snatch a seat from a fellow party candidate than to win a seat from another party. This accurately describes the fact that electoral competition in Denmark is not only an inter-party game but also very much an intra-party affair. The nomination and intra-party seat allocation options now available to parties in the ten MMCs and used for the first time in the 2019 elections are as follows:

1. Nomination by district as mentioned above. This option was only used very rarely in 2019. 2. Parties can decide to combine option (1) with a so-called party list where the party decides the order in which its candidates are to be elected. The list is (almost) a closed list; an extremely high number of preferential votes for a candidate can allow that candidate to break the party list even though it has only happened rarely. This option is primarily used by left-wing parties. In 2019, it

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   69 was only used by the Red-Green Alliance in the nine largest MMCs and by one of the small parties in five MMCs. 3. Parties can also opt for ‘standing in parallel’ (sideordnet opstilling), which is by far the most common option. Here, a party will nominate several candidates— normally, all its candidates in the MMC—in all nomination districts. Parties using this option can now either have their candidates listed alphabetically on the ballot paper or decide to apply some other order. In both cases, they must also decide if the list is to be open, i.e. party votes are disregarded for the intra-party allocation of seats (another novelty in 2019) or only semi-open as was formerly the most popular nomination method (Elklit 2005/2008). The semi-open option includes a distribution of the party votes in each nomination district to all candidates nominated in this way in proportion to their share of the preference votes cast for the party’s candidates in a district. In a few cases, this led to candidates with fewer preference votes overall being elected instead of fellow party candidates with more preference votes overall, which was seen as grossly unfair by the latter. The solution has been to leave it to the parties to choose between the two options. 4. This means that political parties using sideordnet opstilling now have four options: prioritized or alphabetical, and in both of these cases, either open or semi-open. If the prioritized option is chosen, parties must consider if they want to give the top position on the ballot paper in the various nomination districts to their local candidate or not; if they then ‘prioritize’ other candidates in alphabetical order, they are, in effect, using the nomination system used by most parties for most elections since 1970. If readers are confused, they are not alone! But parties and candidates are fast learners when it comes to seat allocation rules, and the 2019 elections saw no major problems caused by the recently implemented system. One reason could be that most voters are not aware of—or interested in—the reasoning behind the different ways in which the parties’ candidates appear on the ballot paper, and they would also normally only see one ballot paper (in the nomination district where they cast their vote), and therefore, not detect that the ballot papers in neighbouring districts are a little different (even though they contain precisely the same candidate names). Figure 5.3 is a copy of a 2019 ballot paper from Nomination District 10 in Copenhagen MMC (Falkoner-kredsen). No matter which particular option an MMC party branch has chosen, when all the votes have been counted and party votes allocated to candidates (where relevant), candidates will be listed in rank order according to the total number of votes they have won. If a party in an MMC has won three seats—whether constituency or compensatory seats— the party’s three candidates with the most votes (only preference votes or preference votes + some party votes) will each obtain a seat in Parliament in declining order, while other candidates will be substitutes, also in declining order of their number of votes. With a list of substitutes available and a clear picking order, there is no need for by-elections. Constituency seats are given out first, but subsequent to that, there is no dis­crim­in­ation of any kind between MPs having a constituency or a compensatory seat.

OUP UNCORRECTED AUTOPAGE PROOFS – , 06/15/2020, SPi

70   Jørgen Elklit

Figure 5.3  2019 ballot paper from Nomination District 10 in Copenhagen MMC. Note: The ballot paper is a long piece of white paper. It has been cut in two here to fit the page format of the book

The Nature of Electoral Law Amendments Electoral law amendments prior to 1961 have been discussed elsewhere, and for space reasons, these discussions are not repeated here (Elklit 1992, 1999, 2002). The 1961 lower­ ing of the electoral threshold is, however, interesting and might provide some insight into the nature of electoral law change in Denmark. The 1953 Electoral Law, which accompanied the approval of major constitutional amendments, provided for an increase in the most important electoral threshold (from approximately 1.1 per cent to about 2.6 per cent) caused by the major parties’ experiences

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   71 with new parties and splinter parties during previous decades. The smaller parties protested, but the Social Democrats, the Liberals, and the Conservatives got an increase enacted, even though the result in 1953 was a softer solution than first suggested. The new threshold opened three doors to parliamentary representation: a constituency seat; 60,000 votes; or that the party had won as many votes as the regional vote/constituency seat average in all three main regions. A new right-wing party (The Independents) did not make it in the September 1953 and 1957 elections even though they were close. The November 1960 elections saw both The Independents and the new Socialist People’s Party succeed in obtaining representation, while one of the opponents of the threshold increase, the Communists, did not clear the hurdle. The 1960 election led to a minority coalition government of Social Democrats and Social Liberals, i.e. one government partner had supported the threshold increase, while the other had been opposing it. The issue was mentioned in the government’s coalition agreement as an issue to be looked into, and despite stubborn opposition from some Social Democrats, the Social Liberals managed to obtain their coalition partner’s support for lowering the threshold through skilful negotiation—and because the Social Democrats wanted to remain in power. The result was a softening of two of the three different ways to obtain representation: the absolute number 60,000 was replaced with a 2 per cent requirement, and the requirement of the vote/constituency seat average was now only required to be fulfilled in two of the three regions, not all three (Ginsburg 1981). These thresholds are still in force, and there is no reason to expect a change (either way) anytime soon. The small parties—potential coalition partners for bigger parties—are very concerned about the threshold requirements, and they would react immediately if a proposal for increasing the electoral threshold was to be presented. A pair of apparently small amendments to the electoral law were enacted in 2017 and implemented in the 2019 elections. They illustrate well the incremental development of Danish electoral legislation at the same time as they demonstrate how a party can take advantage of a particular situation in Parliament. The 2015–19 centre-right coalition minority government relied in general on the support of the Danish People’s Party in the Folketinget. The Danish People’s Party, therefore, could also on occasion expect the support of the parties in government when it wanted some legislation passed. It had been debated for decades whether an option for parties to run completely open lists should be introduced, and some of the parties—including the Danish People’s Party—had put forward such a proposal as a private member’s bill on a number of occasions. In 2017, the idea was again presented by members of the Danish People’s Party, and suddenly, there was unanimous support for the proposal, which was subsequently employed in the 2019 elections by the parties who considered this option an improvement. Already mentioned is the other novelty passed at the same time, namely that the alphabetical ordering of candidates when a party in an MMC runs in parallel in all nom­ in­ation districts can be replaced by a candidate ordering decided by the party. The 2019 elections saw almost all local party branches use this to promote their preferred (local) top candidate or a joint candidate promoted in all the nomination districts (Statistics Denmark 2020).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

72   Jørgen Elklit Although it bears no consequences in the eventual selection of elected candidates, which is solely based on the totality of preference and party votes (if the party allows for party votes to be allocated), the order of candidates on the ballot paper matters—at least to some degree—as candidates higher up on the ballot paper are more visible to the ­voters and will potentially attract more preference votes than other candidates (BlomHansen et al. 2016).

Comparisons with Other Electoral Systems Electoral systems come in different forms and shapes, which forces comparativists to consider whether they should compare apples and oranges or only various apple sorts. Does it, for example, make sense to compare the disproportionality index values for majority and PR electoral systems when majority systems do not aim at achieving a low level of disproportionality? The differences in index values can still be telling, but they are obviously attributable to basic features of the two main categories of electoral systems well known from the literature. Therefore, Table  5.5 only shows some of the differences between the three Scandinavian systems, as they all are two-tier PR systems where the national level is the decisive level. Nevertheless, one finds remarkable differences in the number of MMCs, in the effectiveness of the preferential vote, in whether snap elections can be called or not, in the formal electoral threshold, in how large the share of compensatory seats are,

Table 5.5.  Selected Features of Scandinavian Electoral Systems Denmark

Norway

Sweden

Number of lower-level MMCs 10

19

29

Preferential voting

Of some importance

Possible but no effect

Very modest effect

When parliamentary elections can be called

At any time

Snap elections not possible

Snap elections can be called but only for the remaining part of the term

Main electoral threshold

2 per cent

4 per cent

4 per cent

Share of all seats which are compensatory seats

23 per cent

11 per cent

11 per cent

Gallagher’s index of disproportionality at most recent election

0.024

0.032

0.006

Average of Gallagher’s index at four most recent elections

0.011

0.030

0.018

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   73 and in the value of the disproportionality index at recent elections (Elklit 2016b). The index values not only reflect structural features of the electoral systems but also the number of parties below the relevant threshold in a particular election (vide the values for Denmark and Sweden).

Conclusion Overall, the Danish electoral system functions very well, and the nature and functionality of the electoral system are not regarded as political issues and have not been so for decades. Whether or not it is the best electoral system in the world, as almost suggested by Lijphart, there can be no argument that it is extremely stable, well-functioning, and widely endorsed in Denmark. Since its inception 100 years ago, the electoral system has developed incrementally to accommodate changing demographic, political, and other conditions so that one hears no complaints about its workings. Even politicians from parties that do not make the electoral threshold do not argue that the threshold is unfair or that the election administration is to be blamed for them not making it into Parliament. This—together with the high quality of the work by both central and local election administration staff—provides for a high level of electoral legitimacy. The system’s legitimacy is never challenged, and there has not for many years been serious calls for sim­ plification or greater transparency—even as regards the complexity of the seat allocation procedures created, inter alia, by the parties’ many options with regard to the nom­in­ ation of candidates, list ordering, and computation of which candidates will make it into the Folketinget. This trust in the electoral system is one of several factors which explain why fundamental system changes are never suggested, and ideas of looking into the possibility of introducing some form of digital voting—even only in the form of pilot studies—were shot down immediately. As the argument goes, ‘An ordinary pencil in a ballot booth cannot be hacked!’ However, it is most likely that the system for the submission of signatures in support of a new party will be redesigned before the next election so that the idea of a pensive period during the process cannot be circumvented. This is then yet another example of the incremental development of the Danish electoral system. The 2019 parliamentary elections saw a small increase in the effective number of parties in the electorate to 6.5 and in the effective number of parties in Parliament to 5.9. The value of Gallagher’s disproportionality index also rose to 0.0238 (some would say 2.38) from previous elections’ 0.007 or 0.008. The reason is that three small parties did not make any of the thresholds, and therefore, did not gain any seats. Together, they had 4.3 per cent of the total vote, which meant that the other parties with 95.7 per cent of the vote could share 100 per cent of the seats, i.e. most of them got a slightly larger share of seats than of votes, which contributes to explaining the increase in the disproportionality index. The index is still low, however, compared to most other countries.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

74   Jørgen Elklit Thus, the electoral system—for the 38th consecutive time since 1920—provided Denmark with a trustworthy, fair, and unchallenged outcome to the election.

References Blom-Hansen, Jens, Jørgen Elklit, Søren Serritzlew, and Louise  R.  Villadsen (2016). ‘Ballot position and election results: Evidence from a natural experiment’, Electoral Studies, 44: 172–83. Catt, Helena, Andrew Ellis, Michael Maley, Alan Wall, and Peter Wolf (2014). Electoral Management Design. Revised edition. Stockholm: International IDEA. Elklit, Jørgen (1992). ‘The best of both worlds? The Danish electoral system 1915–20 in a comparative perspective’, Electoral Studies, 11/3: 189–205. Elklit, Jørgen (1999). ‘What was the problem if a first divisor of 1.4 was the solution?’, in Erik Beukel, Kurt  K.  Klausen, and Poul  E.  Mouritzen, eds, Elites, Parties and Democracy. Festschrift for Professor Mogens N. Pedersen. Odense: Odense University Press, 75–101. Elklit, Jørgen (2002). ‘The politics of electoral system development and change: The Danish case’, in Bernard Grofman and Arend Lijphart, eds, The Evolution of Electoral and Party Systems in the Nordic Countries. New York: Agathon Press, 15–66. Elklit, Jørgen (2005/2008). ‘Denmark: Simplicity embedded in complexity (or is it the other way round)?’, in Michael Gallagher and Paul Mitchell, eds, The Politics of Electoral Systems. Oxford: Oxford University Press, 453–71. Elklit, Jørgen (2016a). ‘Valgsystemerne’, in Jørgen G. Christensen and Jørgen Elklit, eds, Det demokratiske system. 4. ed., Copenhagen: Hans Reitzels Forlag, 30–63. Elklit, Jørgen (2016b). ‘Hvor meget styrer folket? En sammenligning af de skandinaviske lande’, in Børge Dahl, Michael H. Jensen, and Søren H. Mørup, eds, Festskrift til Jens Peter Christensen. Copenhagen: Jurist- og Økonomforbundets Forlag, 149–66. Elklit, Jørgen, and Nigel S. Roberts (1996). ‘A category of its own? Four PR two-tier compensatory member electoral systems in 1994’, European Journal of Political Research, 30: 217–40. Gallagher, Michael (1991). ‘Proportionality, disproportionality and electoral systems’, Electoral Studies, 10/1: 33–51. Gallagher, Michael (2005). ‘Indices of fragmentation and disproportionality’, in Michael Gallagher and Paul Mitchell, eds, The Politics of Electoral Systems. Oxford: Oxford University Press, 598–606. Ginsburg, Georg (1981). ‘Slaget om valglovens spærreregler’, Weekendavisen BERLINGSKE AFTEN, 26 June–2 July: 4. Herron, Erik  S., Robert  J.  Pekkanen, and Matthew  S.  Shugart (2018). ‘Terminology and basic rules of electoral systems’, in Erik  S.  Herron, Robert  J.  Pekkanen, and Matthew S. Shugart, eds, The Oxford Handbook of Electoral Systems. New York: Oxford University Press, 1–20. Houlberg, Kurt, and Niels Ejersbo (2020). ‘Municipalities and regions. Approaching the limit of decentralization?’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 141–59. Kjær, Ulrik (2020). ‘Local elections. Localized voting within a nationalized party system’, in Peter M. Christiansen, Jørgen Elklit, and Peter Nedergaard, eds, The Oxford Handbook of Danish Politics. Oxford: Oxford University Press, 382–99.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 06/15/2020, SPi

The Electoral System: Fair and Well-Functioning   75 Laakso, Markku, and Rein Taagepera (1979). ‘ “Effective” number of parties. A measure with applications to West Europe’, Comparative Political Studies, 12/1: 3–27. Lijphart, Arend (1994). Electoral Systems and Party systems. A Study of Twenty-Seven Democracies 1945–1990. Oxford: Oxford University Press. Lijphart, Arend (2005). ‘Foreword’, in Michael Gallagher and Paul Mitchell, eds, The Politics of Electoral Systems. Oxford: Oxford University Press, vii–x. Miller, Nicoline  N., and Jørgen Elklit (2019). Folketingsvalgloven med kommentarer. Copenhagen: Jurist- og Økonomforbundets Forlag. Norris, Pippa, Thomas Wynter, and Sarah Cameron (2018). Electoral Integrity & Campaign Media. The Electoral Integrity Project 2018 mid-year update. Sydney: The EIP Project. www. electoral.integrityproject.com/2018midyearupdate/ (accessed 24 November 2019) Renwick, Alan (2018). ‘Electoral system change’, in Erik S. Herron, Robert J. Pekkanen, and Matthew  S.  Shugart, eds, The Oxford Handbook of Electoral Systems. New York: Oxford University Press, 113–32. Shugart, Matt S., and Martin P. Wattenberg (2001). ‘Conclusion: Are mixed-member systems the best of both worlds?’, in Matt S. Shugart and Martin P. Wattenberg, eds, Mixed-Member Electoral Systems: The Best of Both Worlds? Oxford: Oxford University Press: 571–96. Statistics Denmark (2019). Opgørelse af folketingsvalget den 5. juni 2019. This is the official specification of the election result. Also available at www.dst.dk/valg/Valg1684447/other/ Folketingsvalg2019.pdf (accessed 26 June 2019). Statistics Denmark (2020). Folketingsvalget 5. juni 2019. Danmark, Færøerne, Grønland. Copenhagen: Danmarks Statistik. Will also be available at www.dst.dk/Publ/FolkeTingsValg (accessed 6 November 2019). This link also gives access to electoral statistics from all previous Folketinget elections, i.e. since 1849. Taagepera, Rein, and Matthew S. Shugart (1989). Seats and Votes. The Effects and Determinants of Electoral Systems. New Haven, CT: Yale University Press. The election-related homepage of the Ministry of Social Affairs and the Interior is available at https://valg.sim.dk; an English version is available at https://elections.sim.dk (both accessed 6 November 2019). The Folketinget electoral law is available in English at https://elections.sim.dk/media/21968/ folketing-parliamentary-elections-act-2019.pdf (accessed 6 November 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

chapter 6

Elector a l T u r nou t Strong Social Norms of Voting Kasper M. Hansen

Very High and Stable Turnout Electoral turnout is the most important health indicator for democracies with a non-compulsory voting system. Voting holds representatives accountable for their deeds and grants a mandate to rule in the next term on behalf of the people. But the act of voting is also an active expression of consent for the representative political system that provides democratic legitimacy to the system (Rousseau 1762; Beetham 1991). If turnout is low, democratic legitimacy is weak, and if turnout declines, so does democratic legitimacy, the foundation of the democratic system (Lijphart 1997; Wolfinger and Rosenstone 1980). If one defines turnout as the share of the electorate that votes, it becomes a very useful indicator of a healthy democracy that can be compared across time and space. The four research questions that guide the chapter are: What is the turnout level in Denmark? Why is it so high comparatively? Which inequalities exist between turnout groups? And lastly, can Get-Out-The-Vote (GOTV) campaigns increase turnout equality in a high-turnout country like Denmark? Voting is non-compulsory in Denmark, and there is a very effective automatic voter registration system based on the universal civil registration number granted by birth or residence in Denmark. This means that all eligible persons automatically receive their polling card by regular mail before the election. For national parliament elections, one is eligible to vote if one is a Danish citizen, 18 years of age, and has permanent residence in the country (Christensen 2020). For other elections, the electorate also includes EU and Nordic citizens with permanent residence. In addition, other citizens are eligible for regional and municipal elections if they have had permanent residence in Denmark for at least 3 years. Denmark has experienced high and stable turnout rates over recent decades (Figure 6.1), also compared to other countries (IDEA 2019).

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

Electoral Turnout: Strong Social Norms of Voting   77 100 90 80 Turnout (pct.)

70 60 50 40 30 20 10

Municipal elections European Parliament

2019

2015

2011

2005 2007

2001

1998

1994

1987 1988 1990

1984

1970 1971 1973 1975 1977 1979 1981

0

County/regional elections National Parliament

Figure 6.1  Turnout in Denmark 1979–2019, per cent.

The Parliament (Folketinget) elections in particular enjoy high and stable turnout, averaging 86 per cent since 1970. The municipality elections’ average turnout is 70 per cent—if the 2001 municipality elections are excluded as they are a strong outlier due to the simultaneously held national election. To compare, the European Parliament elections’ average turnout is only 53 per cent. Since the first Folketinget election where women had the vote (1918), the turnout rate for Folketinget elections has never been below 75 per cent. This is impressive, also considering that the voting age was lowered several times during the twentieth century, expanding the number of eligible voters considerably. The high turnout level in Denmark also corresponds to a high level of satisfaction with democracy and a high level of trust in government and politicians, even though the latter has decreased over the latest decade (Bengtsson et al. 2014: 44; Stubager et al. 2019). These relationships point to high turnout at least going hand-in-hand with other more subjective measures of a well-functioning democracy. In the 2014 European Parliament elections, 55 per cent of Danish voters declared that voting for that particular election was a duty, placing Denmark in the top 5 of the 28 EU countries (Schmitt et al. 2016). This strong sense of duty is one of the key explanatory factors in trying to understand a high turnout level (Blais 2000), but quite often, it just raises the next question: where is the high sense of duty emerging from? Research points in this regard to a country’s strength of democracy and social homogeneity (Galais and Blais 2017). We could add to this a strong democratic culture that is nursed in primary education and through a strong civil society. Many of the civil society organizations in Denmark still have links to

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

78   Kasper M. Hansen the political parties and the broader public today, which contribute to a strong political awareness among the public. In addition, the mobilization of the electorate through the labour movement must not be neglected. Denmark has Europe’s highest unionization rates, with close to 70 per cent of all employees being union members (ETUI 2018: 61). Even though the unions’ ties to the political parties have been weakened in recent decades, the labour unions still contribute to the ongoing mobilization of the voters (Elklit and Togeby 2009; Henriksen et al. 2018). Another factor contributing to the high turnout in Denmark is the early establishment of a popular right-wing party. It was first established in 1972 as the Progressive Party and later transformed to the Danish People’s Party. This party has been able to mobilize voters on the popular right that otherwise would have been likely to abstain (Elklit and Togeby 2009). This conclusion is also supported by the fact that the party finds its base among the least educated and least interested in politics (Andersen 2017; Hansen 2007). Another important feature that contributes to the high electoral turnout in Denmark is the strong ritual and traditions around voting (Hansen et al. 2017). Most people vote by showing up at the polling station on election day, and only few vote absentee in advance. In the 2015 and 2019 parliamentary elections, only 8.7 and 8.4 per cent of all participating voters cast an advance vote. Quite often, people vote together with fellow housemates and make it a joint social venture to go and vote. In the 2015 parliamentary election, 56 per cent of all voters in households with more than two eligible voters voted at the same time as someone else in the household (Bhatti et al. 2020). The effective automatic voting registration, the strong feeling of citizen duty developed over previous decades, strong civic society movement, the early mobilization of the popular right, and the social aspect of voting all contribute to the explanation of the high turnout in Denmark. Finally, it is worth mentioning that the strong public service media in Denmark, which to a large extent dominate the news coverage, cover the elections extensively and often directly encourage viewers and listeners to participate in the election.

Who Votes? Not only is the level of turnout important for assessing the level of democratic le­git­im­ acy, so is the equality in turnout across groups. If certain groups have low or declining turnout, their voice is less heard in the political system, and policy developments might be less responsive to the group’s wishes (Griffin and Newman 2005; Leighley and Nagler 2013). Furthermore, such groups might find it even more difficult to identify with the legislators, develop mistrust, etc. (Mansbridge 1999; Bhatti et al. 2018). In Denmark, it has been possible to collect actual and validated turnout data since 1997. Across elections, it has been possible to merge the turnout data for almost the entire population with Statistics Denmark’s several hundred socio-demographic vari­ables, which have allowed the most fine-grained analyses of turnout in the world (e.g. Hansen 2018). These analyses have fundamentally revised our understanding of why people vote.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

Electoral Turnout: Strong Social Norms of Voting   79 It has long been known that there is a general curvilinear relationship between turnout and age as the young and the elderly vote less than other age groups. But the Danish turnout data have allowed a more fine-grained analysis which has revealed the ‘rollercoaster’ relationship between age and turnout. First of all, the analysis has revealed a sharp decline from 18 years of age only until 21 years of age, which is not visible with sample data, self-reported turnout data, or summarized analyses with age groups collapsed (Bhatti and Hansen 2012a; Bhatti et al. 2012; Bhatti et al. 2016a, 2016b; Dahlgaard et al. 2019). The general explanation of this decline is that when the youngest leave home, their turnout declines rapidly as their social ties are disrupted. This picture is now considered universal and confirmed in the United States, Norway, Sweden, and Finland where data are available (Bhatti et al. 2012). Figure 6.2 depicts the relationship between age and turnout in the 2015 parliamentary election and the local election in 2017. A second important observation is that the rollercoaster turnout peaks among senior citizens around the age of 72 and then declines sharply. This decline cannot be observed when survey data are used. The decline among senior citizens is partly due to deteriorating health and retirement from work, but it is also related to the social aspect of voting as more live alone in old age, especially women. The difference in turnout between genders is also notable. Young women have a substantially higher turnout than young men. This is partly explained by lifecycle patterns as women start education, family, and the like earlier than men, but also by the fact that women on average today obtain longer educations than men. Men over 72 years have

100

Turnout (pct.)

80 60 40 20 0 18 21

30

40

60

50

70

80

90

100

Age in years

Men, Local Men, National

Women, Local Women, National

Figure 6.2  The rollercoaster: Turnout across age and gender in the 2015 parliamentary e­ lection and the 2017 local elections. Note: Actual validated turnout based on administrative data; dashed horizontal lines are average turnout in the two elections

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

80   Kasper M. Hansen 100

Turnout (pct.)

80 60 40 20 0 18 21

30

40

50

60

70

80

90

100

Age in years Danes, Local Danes, National

1st gen. non-western imm. 1st gen. non-western imm.

2nd gen. non-wes. imm. 2nd gen. non-wes. imm.

Figure 6.3  Turnout across ethnicity and age in the 2015 parliamentary elections and the 2017 local elections. Note: Actual validated turnout based on administrative data; dashed horizontal lines are average turnout in the two elections

higher turnout than women, but this is explained by the fact that as women live longer and marry older men, women will also live longer alone. Adjusting for living alone makes the difference in old age across gender disappear (Bhatti and Hansen 2012b). Another point is that the rollercoaster relationship is even steeper when we analyse less salient elections, which then confirms Tingsten’s (1937) law of dispersion that stipulates that equalities in turnout across groups increase when turnout declines (Bhatti et al. 2018a). This is also clear from comparing national and local elections in Figure 6.2. Another notable inequality in turnout is among different ethnic groups. Figure 6.3 shows the turnout across age and ethnicity. First-generation and second-generation immigrants from non-Western countries have substantially lower turnout than Danes. The difference is somewhat larger in the local elections than in the national elections as expected according to Tingsten’s law of dispersion. It is troubling from an equality standpoint that especially second-generation immigrants—who are born in Denmark—have not become active voters and are still close to the turnout level of the first-generation immigrants (i.e. their parents) as it suggests that the democracy component of integration has not succeeded. It is, therefore, a challenge to find ways to mobilize immigrants to participate actively in Danish democracy at the same level as Danes. As this group will increase in size relative to Danes in the coming years, one should expect a gradual decline in turnout. We can analyse the turnout for specific groups in even more detail. Table  6.1 ­presents the turnout for particular groups in the 2015 national election and the 2017

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

Electoral Turnout: Strong Social Norms of Voting   81

Table 6.1.  Turnout across Specific Groups, 2017 Local and 2015 National Elections Groups

Local 2017 (pct.)

National 2015 (pct.)

60–69 years old Long-term tertiary education DKK 400.000–499.999 yearly income Short and medium-term tertiary education DKK 300.000–399.999 yearly income 50–59 years old Danish ethnicity Vocational education 40–49 years old Women Men DKK 100.000–199.999 yearly income Secondary education (e.g. high school) Primary school DKK 200.000–299.999 yearly income 30–39 years old Less than DKK 100.000 yearly income 22–29 years old Immigrants, Danish citizens Early retirement Immigrants from Iceland and Norway Immigrants from old EU countries Immigrants from other Western countries Unemployment benefits Immigrants from non-Western countries Second-generation immigrants First-generation immigrants, non-Danish citizens First and second-gen. imm. from new EU countries

81.1 80.5 80.5 80.0 75.6 75.3 75.2 73.3 72.6 72.4 68.8 67.5 66.7 66.7 65.9 63.7 57.1 54.9 54.7 53.8 47.7 46.4 46.2 43.5 42.0 40.9 31.8 18.5

91.1 95.4 92.9 93.2 90.0 88.6 87.0 87.5 87.7 86.9 84.7 82.1 86.1 76.7 82.1 85.0 77.7 79.1 66.1 69.7 82.3 82.7 76.8 62.9 58.6 53.4 Not eligible 72.6

Note: Old EU countries are the first 15 member states. New EU countries are EU27–EU15, excluding Croatia. Iceland and Norway have their own category. Other Western countries are Lichtenstein, Monaco, San Marino, Andorra, Switzerland, Canada, USA, Australia, and New Zealand. Non-Western countries are all other ­countries. The dotted horizontal line indicates whether the groups are above or below turnout average in the particular election.

local election. It is obvious that there are substantial turnout inequalities between the groups identified in the table. The groups with the highest turnout are the 60–69-year-olds and voters with higher education. At the bottom of the list, immigrants from new EU countries and from nonWestern countries dominate. If we combine the groups’ characteristics, the difference between high and low turnout becomes even more profound. For example, the group

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

82   Kasper M. Hansen consisting of retired 60–69-year-old women with higher education and Danish citizenship had a turnout of 93.4 per cent in the 2017 local elections. This can be compared to the group of 22–29-year-old immigrants from new EU countries without Danish citizenship. The turnout in this group in the same election was only 4.8 per cent. We can also analyse turnout inequalities by comparing core voters and permanent abstainers across the different types of elections in Denmark. If we look at voters that voted in the three consecutive elections consisting of the national parliament elections in 2015, European Parliament elections in 2014, and the local elections in 2013, we find that the group is composed of 52 per cent females, 1 per cent non-Western immigrants, and 39 per cent with higher education. If we compare this group to the group that did not vote at all in the same three elections, its composition is 47 per cent females, 12 per cent non-Western immigrants, and only 11 per cent with higher education. So there are more than 10 times as many non-Western immigrants in the permanent abstainers’ group as in the group of core voters and more than 3 times as many highly educated in the core voter group compared to the permanent abstainer group (Bhatti et al. 2018a). We can describe these two groups, i.e. core voters and permanent abstainers, as a democratic A-team dominated by highly educated females that vote in every election compared to a democratic B-team dominated by men with low levels of education and a non-Western, immigrant background. This clearly illustrates the large inequalities in the various groups’ turnout even in a country like Denmark with a high general turnout.

Denmark: A European Frontrunner in Get-Out-The-Vote (GOTV) The lower turnout in elections to the European Parliament and in local elections has created some interest in trying to mobilize voters in Denmark. There has, in particular, been an interest in the groups with low turnout such as young voters and ethnic mi­nor­ ities, but many other low-turnout groups have also been targeted in these mobilization campaigns. Some of the efforts have been conducted as large field experiments with randomly controlled trials building on the American tradition (Green and Gerber 2015). Various modes of delivery have been tested (postcards, letters, e-mails, digital mails, text messages on cell phones, door-to-door, and posters in hallways in social housing), the timing of mobilization (distance to election day and time of day), number of mobilization efforts, and the content of the mobilization (the Constitution, cartoons, focus on the duty of voting or the social aspect, etc.) (Bhatti et al. 2015, 2017a, 2017b, 2018b). What can be learned from these experiments? First, it is possible to increase turnout even in a high-turnout country. Turnout among the youngest voters has actually increased more than 17 percentage points from 2009 to 2017 at local elections, which is probably a world record.

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

Electoral Turnout: Strong Social Norms of Voting   83 Nevertheless, only a small part of this mobilization effect can be directly contributed to the GOTV field experiments as the effect size of these experiments are similar to US GOTV experiments which are typically less than 2 or 3 percentage points (Bhatti et al. 2019; Green et al. 2013). The GOTV field experiments have also successfully identified a direct spillover effect of 30 per cent to other household members, even parents. Thus, the mobilization efforts travel not only downstream from parent to young voters but also in the opposite direction from the young voter to their parents (Bhatti et al. 2017c). Table 6.2 provides an overview of the effect size on turnout from our GOTV field experiments. It includes both the most successful experiments and the ones which did not have any mobilizing effect in order to provide a fuller picture of the experiences across local and European Parliament elections. So even though the many field experiments have provided solid evidence of their effect using randomized controlled trials, it is also evident that the effect sizes are relatively modest compared to the structural aspect of the differences in turnout, such as the difference across age, gender, and ethnicity. Thus, GOTV undertakings are in no way a quick fix to reducing the substantial differences in turnout among the various groups of voters. Nevertheless, the experiments also have an indirect effect as they have increased the general awareness about turning out to vote from the media and public. For ex­ample, many media outlets took special initiatives to increase turnout in both 2013 and 2017 by devoting part of their election coverage to the aspect of turning out to vote.

Table 6.2.  Overview of the Experiment’s Total Effects Experiment/group

Effect size in percentage points

Election and year

Posters and flyers, Bolbro, Odense municipality (imm.)

14.4 *

Local 2013

Postcards, Silkeborg municipality

2.9 *

Local 2013

SMS, Ministry for CGEIS* (second-gen. immigrants)

2.9 *

Local 2013

Letter and postcard, Silkeborg municipality (joint effect)

2.8 *

Local 2013

Letter, Silkeborg municipality

2.8 *

Local 2013

SMS, Danish Youth Council (before election day)

2.3 *

Local 2013

SMS, Danish Youth Council (joint effect)

1.8 *

Local 2013

Letter, Danish Parliament (cartoon and Constitution)

1.6 *

Local 2013

SMS, Danish Parliament (22-29 years old)

1.3 *

Local 2013

Constitution, Danish Parliament (joint effect)

1.1 *

Local 2013

Letter, Ministry for Eco. Affairs & Interior (all arguments)

1.1 *

Local 2013

SMS, Council for Ethnic Minorities

1.1 **

Local 2017

SMS, Ministry for CGEIS* (immigrants)

1.0 *

Local 2013 (continued )

OUP CORRECTED AUTOPAGE PROOFS – FINAL, 05/19/2020, SPi

84   Kasper M. Hansen

Table 6.2.  Continued Experiment/group

Effect size in percentage points

Election and year

SMS, Danish Youth Council (on election day only)

0.8

Local 2013

Door-to-door, Social Democrats (joint effect)

0.7

Local 2013

SMS, Danish Youth Council (joint effect)

0.7 *

EU 2014

Constitution, cartoon, Danish Parliament

0.7 **

Local 2017

SMS, Ministry for CGEIS* (without link)

0.6 *

Local 2013

Letter, Danish Parliament (letter and Constitution)

0.6

Local 2013

Posters in (social housing), Council for Ethnic Mino.

0.6

Local 2017

SMS, without name, Danish muni. (40+ age group)

0.5 *

Local 2017

Letter, Min. for Eco. Affairs & Interior (joint effect)

0.4 *

Local 2013

SMS, Danish Parliament (joint effect)

0.4

Local 2013

SMS, Ministry for CGEIS* (joint effect)

0.3

Local 2013

Digital letter, Silkeborg municipality

0.3

Local 2017

Postcard, Silkeborg municipality

0.1

Local 2017

SMS, Ministry for CGEIS* (with link to VAA)

0.0

Local 2013

SMS, with name, Danish municipalities (40+ years group) −0.1

Local 2017

Posters and flyers, Bolbro, Odense muni. (joint effect)

−0.3

Local 2013

SMS, Ministry for CGEIS* (ethnic Danes)

−0.4

Local 2013

E-mail, DaneAge Association (joint effect)

−0.4

Local 2013

Door-to-door, We vote together, CPH Uni. (joint effect)

−0.8

Local 2013

Door-to-door, Voting on the edge, Huset Zornig**(joint eff.) −1.9

Local 2013

Posters and flyers, Bolbro, Odense muni. (ethnic Danes)

−2.4

Local 2013

Door-to-door, Local union, Randers muni. (joint effect)

−3.8

Local 2013

Note: * p